15 th International Open and User Innovation Conference. Book of Abstracts. July 10 12, 2017 Universität Innsbruck

Transcription

1 15 th International Open and User Innovation Conference Book of Abstracts July 10 12, 2017 Universität Innsbruck

2 OUI 2017: Conference Topics and Chairs Topics Communities: User Innovation and Open Source Contests, Crowdsourcing Open Innovation Crowdfunding and Firm s Interaction with User Innovation Law and IP Innovation Policy Problem Solving and Toolkits User Innovation and Diffusion User Innovation and Psychology User Innovation and Healthcare Sharing Economy and Platforms Chair(s) Christopher Lettl, Vienna University of Economics & Business Joachim Henkel, Technical University of Munich Johann Füller, Universität Innsbruck Frank Piller, RWTH Aachen University Johann Füller, Universität Innsbruck Ronald Maier, Universität Innsbruck Lars Bo Jeppesen, Copenhagen Business School Christina Raasch, Technical University of Munich Katherine Strandburg, New York University School of Law Andrew Torrance, University of Kansas, School of Law Peter Svensson, The Institute of Technology at Linköping University Dietmar Harhoff, Max Planck Institute for Innovation & Competition Eric von Hippel, Massachusetts Institute of Technology Georg von Krogh, ETH Zurich Nikolaus Franke, Vienna University of Economics and Business Jeroen de Jong, Utrecht University School of Economics Matthew O Hern, University of New Hampshire Steven Flowers, Kent Business School, University of Kent Harry DeMonaco, Católica-Lisbon School of Business & Economics Pedro Oliveira, Católica-Lisbon School of Business & Economics Wouter Boon, Utrecht University, Innovation Studies Group

3 Table of Content Communities: User Innovation and Open Source Paper Designing Organizations for Productive Bursts Georg von Krogh, Thomas Maillart, Stefan Haefliger, Didier Sornette 2 Social Identification and Entrepreneurial Action: An Investigation into Hacker- Maker Communities Maria Halbinger, Francesca Melillo 3 Triggers of Collaborative Prototyping: A Netnographic Study of User Innovation in Open Source Hardware Communities Matti Grosse, Jakob Pohlisch, Jakob J. Korbel 4 Who Moves to the Centre of Gravity? Developers Progression from Periphery to Core in the Linux Kernel Development Project Marvin Hanisch, Stefan Berreiter, Carolin Häussler, Sven Apel 6 Open Radar Groups: The Integration of Online Communities into Open Foresight Processes Michael Andreas Zeng, Hans Koller, Jahn Reimo 7 Three Triggers of Search Why Young Firms Change Selective Revealing over Time Joachim Henkel, Dominik Hepp 8 Poster Legitimacy Creation in Grassroots Innovations: Community Supported Agriculture in the Netherlands Ellen H.M. Moors, Laura M. van Oers 9 Delegation in Knowledge Creating Online Communities Shiko Ben- Menahem, Yash Raj Shrestha, Georg von Krogh 12 Combining Crowdsourcing and Toolkits to a Business Model: The Case of Stata Corp. Kathrin Reinsberger, Nikolaus Franke 15 Teaching User Innovation through Competition The challenge of crossfunctional teams in digital learning environment Minna-Maarit Jaskari, Mona Enell- Nilsson, Jussi Kantola 18 Innovating beyond Firm Boundaries: Effects of Resource Deployment Control in Open Source Software Development Mario Schaarschmidt 20

4 Open Source Software Developer-Community Role Conflict and Organizational Turnover Mario Schaarschmidt 22 Contests, Crowdsourcing and Open Innovation Paper An Organization with many Faces: A Crowd-Based Venture s Versatility in Resource Mobilization Thomas Gegenhuber, Robert M. Bauer 26 The Double Selection Environment: Ideas Selected by Crowds and Experts Lars Frederiksen, Michaela Beretta, Dirk Deichmann 27 From Few to Many: Scaling Crowdsourcing Design Platforms Thomas Kohler, Lea Rützler 28 Too Many Cooks Spoil the Broth: Community Feedback and Idea Quality in Idea Contests Isabella Seeber, Daniel Zantedeschi, Anol Bhattacherjee, Johann Füller 29 How Robust are the Results? A Bayesian Averaging Approach for Tackling Replication and Model Uncertainty in Research on Inbound Open Innovation Bernd Ebersberger, Fabrice Galia, Keld Laursen, Ammon Salter 30 Entrepreneurial Ventures as Sources for Innovation? How Incumbents (Under)Utilize Corporate Incubators and Accelerators as an Open Innovation Approach Sandra Luisa Moschner, Cornelius Herstatt 31 Mapping Open Innovation in SMEs and Large Companies in KSA: A Case in the Industrial City in Riyadh Amal Al- Dulaigan, Afaf Bugawa, Odeh Al- Jayyousi 33 Attenuating the Not-Invented-Here Syndrome: The Value of Behavioral Remedies Julian Hannen, David Antons, T. Oliver Salge, Frank Piller 34 Enabling Community-based Crowdsourcing: The Role of Open Innovation Intermediaries and Co-creation Capabilities Krithika Randhawa, Ralf Wilden 35 Crowdsourcing without Profit: How Seeker Motivation and Strategies Drive Local Community Innovations Krithika Randhawa, Ralf Wilden, Joel West 36 Pressure in Crowdsourcing Contests Jonas Heite, Karin Hoisl 37

5 Challenges and Capabilities to Profit from Customer Co-creation in Organizations Christiane Rau, Katja Krämer, Kristiana Pavlova 39 Is Open Innovation in Startups DNA? Zeljiko Tekic 40 Unfolding Agile Innovation Capabilities in Co-Located Innovation Labs Florian Fecher, Johanna Winding, Katja Hutter, Johann Füller 41 B2B Crowdsourcing for NPD and Innovation: A Literature Review and Research Directions Sylvia Dimitrova 42 "Consumers hijacked my idea contest! Managers preference formation and reactions towards deviant content in ideation contests Alexandra Gatzweiler, Vera Blazevic, Frank Piller 42 Poster Organizing for Open Service Innovation: Moving Beyond Manufacturing Deborah Roberts, Simona Spedale, Wolfgang Gruel 45 Gender Imbalance Online Andrea Blasco, Karim R. Lakhani, Michael Menietti, Nidhi Chaudhary 48 The Positive Side of the Corruption Effect: Inducing Controlled Motivation to Increase Accuracy in Crowdevaluation Christian Garaus 49 Self-Selection in Online Innovation Contests Nikolaus Franke, Philipp Topic, Ines Reith 52 Individual Search and Innovation Performance in Open Innovation Anne Greul Tim Schweinsfurth, Christina Raasch, Chia- Huei Wu 55 Factors Influencing the Quality of Open Ideas: Open Innovation, Networks and Interlocking Ties Nassim Belbaly, Rajaa El Mezouaghi, Calin Gurau 56 Divide and Conquer: Experimental Analysis of Mechanisms that enable Crowd Filtering of Crowd-Generated Ideas Viktoria Banken, Isabella Seeber, Alexander Merz, Ronald Maier 57 How Close are Open Innovation Partners? Degrees of Institutionalization of Firm-Consumer Collaboration for Innovation Eva Lendowski, Ansgar Buschmann, Gerhard Schewe 60

6 Investigating Self-Selection Mechanisms Induced by Selective Revealing Markus Deimel, Christopher Lettl 63 Ecosystems in Idea Evaluation. An Explorative Analysis of Multiple Crowdsourcing Platforms Ralph Lichtner 66 Idea-Related Knowledge Accumulation in Online Innovation Collectives Brita Schemmann, Maryse M. H. Chappin, Andrea M. Hermann, Gaston J. Heimeriks 68 Evaluating Crowd-Sourced Ideas: Influence of Idea Characteristics on Crowd and Organizational Evaluators Lisa Kristina Wimbauer, Patrick Figge, Carolin Haeussler 70 Solver Motivations in Complex Innovation Contests Ademir Vroljik 73 The Role of the Entrepreneur in Managing Crowdsourcing in SMEs Izabella Bereczki 75 Who are Your Design Heroes? Exploring User Roles and Behavior in a Community-Based Design Contest Manuel Moritz, Tobias Redlich, Jens Wulfsberg 76 Crowdfunding Paper Beyond the Platform - Designing Enterprise Crowdfunding to Foster Intrapreneurship Robert Kleinscheck, Katja Hutter, Johann Füller 80 Does Hierarchical Distance Bias Idea Evaluation? Tim G. Schweisfurth, Michael A. Zaggl, Claus P. Schöttl, Christina Raasch 81 Poster Crowdfunding Science? Lars Frederiksen 82 Crowdfunding for Family Firms in a Resource Exchanging Context Martin Danler 83 Beyond Funding Exploring the Effect of Enterprise Crowdfunding on Employee Engagement Carina Benz, Niels Feldmann 86

7 Do Consumers Really Care About Crowdfunding? Exploring the Crowdfunding Effect on Innovation Perception Christian V. Baccarella, Timm F. Wagner, Kai- Ingo Voigt 88 Firm s Interaction with User Innovation Paper Wicked Design for Wicked Goals - How Hyperloop Transportation Technologies Challenges Organisation Design Theory Ann Majchrzak, Terri Griffith, David Reetz, Oliver Alexy 92 An Exploration of On-Demand Innovation Model in Organizations Jiang Huang, Jin Chen 93 The Digital Revolution, 3D Printing, and Innovation as Data Aric Rindfleisch, Matthew O Hern, Vishal Sachdev 94 Linking User Innovation and Social Innovation - An Explorative Case Study on Lead User Identification in the Humanitarian Sector Daniel Kruse, Moritz Goeldner, Cornelius Herstatt 95 Universities as Lead Users Sofia Patsali, Stefano Bianchini, Patrick Llerena 96 Poster Customization in the Smart Product Age -Consumer s Response to Smart Products for Product Customization in the Usage Stage Ning Wang, Frank Piller 97 The Prototype Used in the Implementation of the Need-Solution Pairs - The Prototype Use of Single-User and Multi-User to Relate Need-Solution Pairs Akimitsu Hirota 100 How Learning Effects Lead to Higher Quality and More Novel Solutions in Pyramiding Search Nikolaus Franke, Barbara Mehner, Kathrin Reinsberger 101 The Management of User Driven Innovation Processes: Towards a Systematic Review Guido Bortoluzzi, Khatereh Ghasemzadeh 104

8 User Knowledge Utilization and Distributed Sensemaking Andreas Benker 107 When the Use of Open and User Innovation Methods is effective: The Moderating Effect of the Organizational Setup Michael Nobis, Nikolaus Franke, Peter Keinz 109 Law and IP Paper The Strategic Coupling of Intellectual Property Management and Co-Creation Anja Tekic, Kelvin Willoughby 112 The Interplay between User Innovation, the Patent System and Product Liability Laws: Policy Implications Stijepko Tokic 113 Users, Patents and Innovation Policy Katherine Strandburg 114 Poster Open-Source and For-Profit Innovation in the World of Things. A Pilot Study in the Circular Economy in Flanders (Belgium) Geertrui Van Overwalle, Lodewijk Van Dycke 115 Innovation Policy Paper Policy Choices in Supporting Collective Innovation: A Study of Pharmaceutical R&D Consortia Joel West, Paul Olk 119 Identification and Classification of User Innovation: Implications for Future Research Jakob J. Korbel, Matti Grosse 120 Trajectories of Local Open Government: Investigating Managerial Perception of Innovativeness Lisa Schmidthuber, Dennis Hilgers 121 Innovation in the Household Sector: Definitions, Statistical Measurement and Policy Fred Gault 122

9 Open Social Innovation Dynamics and Impact: Exploratory Study of a Fab Lab Network Thierry Rayna, Ludmila Striukova 123 Poster Translating User Innovation Research to Innovation Policy Action Jari Kuusisto, Liting Liang 124 Problem Solving and Toolkits Paper Selective Broadcast Search Improving Problem-Solving s Efficiency Using a Company s Internal Crowd Christian Pescher, Katja Hutter, Johann Füller, Michael Heiss 128 Value Development During the Self-Design Process: A Demonstration and Explanation of the Swoosh Effect Nikolaus Franke, Franziska Metz, Moreau Page 129 Problem Solving Without Problem Formulation: Documenting Need-Solution Pairs in a Laboratory Setting Ruth Stock, Christian Holthaus 131 Scenario Design: Using Design Thinking in Scenario Technique Among Technology Clusters Jahn Reimo, Hans Koller, Michael Andreas Zeng 132 Gaining Insight for Innovation Peter Hu, Shannon Heald, Peter Malonis, Howard Nusbaum 134 Poster Everyone is a designer Radical Innovation to Foster User Creativity in Toolkits Filippova Evgeniia, Nikolaus Franke 135 Application of Structured Methods for Understanding Problems by User Innovators Jennifer Otitigbe 137

10 User Innovation and Diffusion Paper Diffusion of User Innovation Within Online Communities from The Social Network Perspective Wonho Lee, Youngbae Kim 141 Why New Products Appear in Unexpected Places and What We Can Learn from It to Spur Product Development Konstantin Fursov, Jonathan Linton 142 The Field of External Search and the Search for External Knowledge Sara Heuschneider, Daniel Ehls, Cornelius Herstatt 143 Decision Makers Underestimation of User Innovation Philip Bradonjic, Nikolaus Franke, Christian Luethje 144 User Innovation A Systematic Literature Review from 1970 to 2013 Sandra Jennina Sanchez, Alazne Mujika- Alberdi 145 Innovation Diffusion via Remixing Digital Objects: The Role of Word-of-Mouth Commentary in Online User Communities Gregory Fisher, Michael Stanko 146 Opportunity costs of Household Sector Innovation A study in the Emirates Daan Rademaker, Jeroen De Jong 146 Poster User Innovation and Diffusion in Online Firm-hosted User Communities The Case of China Yu- Shan Su 148 Measuring Innovation Performance with Creativity between Free Closed and Open Innovators André Witzel 151 User Innovation and Psychology Paper The Company or the Crowd? Comparing Consumers Reactions to Peer-Provided and Firm-Provided Customer Support Lan Jiang, Matthew O Hern, Sara Hanson 154

11 Poster Uncovering the Value Basis of User Innovation Helle Alsted Søndergaard, John Thøgersen 155 Lead Userness and Innovative Work Behavior in Application Development: A Dual Path Model Mario Schaarschmidt, Dirk Homscheid, Björn Höber, Matthias Bertram 158 The Interplay of E-Lancers Character Traits and Digital Signals Performance Implications in Online Labor Markets Ruth Stock, Christian Holthaus 159 Does Job-related Innovation Benefit User Innovations? An Investigation of Spillover Effects Ruth Stock, Carmen S. Lukoschek 160 From User-Innovator to User-Entrepreneur: Designing and Delivering a University Course to Foster the Co-Creation of Business Models Albrecht Karlusch, Kathrin Reinsberger, Wolfgang Sachsenhofer 163 User Innovation and Healthcare Paper A Next Generation for Public Health Intervention Models: The Public as Innovators Christina von Hippel 167 People with Disabilities as Product Innovators: A Pilot Study Peter Conradie, Aron-Levi Herregodts, Lieven De Marez, Jelle Saldien 168 If necessity is the mother of invention, are patients from developing countries keen innovators? Pedro Oliveira, Helena Canhao, Salomé Azevedo, Joao Silva 168 The Evolution of Stealth Innovation in Nursing: History, Drivers and Prototyping Genome from the MakerNurse Study Anna Young, Nikolas Albarran, David Marshall, Maureen DeMenna, Max-Philipp Schrader, Jose Gomez- Marquez 168

12 Poster The Hurdles to Diffusion of a User Innovation in a Market Dominated by Experts The Case Of PEARS Leid Zejnilovic, Pedro Oliveira 171 The Role of Healthcare Professionals in the Diffusion of Patient Innovations: An Experiment in the Field of Physical Medicine and Rehabilitation Merle Schlottmann, Pedro Oliveira 174 How Lead Users Impact the Decision-Making in Innovation Process? Senda Belkhouja, Corine Genet, Vincent Mangematin 177 The Impact of User Innovation on Patients Health-Related Quality of Life An Explorative Case Study on Medical App Developers Moritz Goeldner, Cornelius Herstatt 179 Patientube - Supporting Patients and their Relatives by Curated Expert and User Generated Peer Videos An Analysis of Critical Success Factors Andrea Hofmann- Rinderknecht, Andreas Kreimaier, Gerhard Buchegger, Dominik Walcher 181 Sustainable Healthcare: Mechanisms of Co-Innovation by Patients and Providers Adam Seymor 183 Sharing Economy and Platforms Paper Explaining the Vertical to Horizontal Transition in the Computer Industry Carliss Y. Baldwin 186 Platform Ecosystems: How Developers Invert the Firm Geoffrey Parker, Marshall Van Alstyne, Xiaoyue Jiang 187 Who Should Benefit in a Sharing Economy Model: A Look at Resource Endowment and Trust Christopher Smolka, Christoph Hienerth, Franz W. Kellermanns 188 To share or not to share Exploring the Impact of Sharing Behaviour on User Innovativeness Frank Tietze, Thorsten Pieper, Carsten Schultz, Cornelius Herstatt 189

13 Open Innovation in the Digital Age. New Options to Close the Gap between Universities and Companies? Boris Alexander Becker 190 Poster Principles of Crowd-Based Organizing: Unpacking Platforms Functions and Processes Robert M. Bauer, Thomas Gegenhuber 192 The Free Strategy : Economics of Open Versus Proprietary Designs Alfonso Gambardella, Eric von Hippel 195 Pay-As-You-Drive Models in The Sharing Economy: A Comparison of German and U.S. Car Owners Mario Schaarschmidt, Raoul Könsgen, Björn Höber, Patrick Hacker 196 The Role of Users in the Platform Economy A Multiple Case Study of Institutional Change by Users of Airbnb Wouter Boon, Kristy Spruit 198

14 Communities: User Innovation and Open Source 1

15 Designing Organizations for Productive Bursts Paper Georg von Krogh (ETH Zurich) Thomas Maillart (University of Geneva) Stefan Haefliger (Cass Business School) Didier Sornette (ETH Zurich) An increasing number of firms across various industries adopt the methods of open innovation, and so extensively share ideas, knowledge, and technology with individuals and organizations outside their boundary. Open innovation brings many benefits to firms, such as lower cost of product development. Yet, for management it may create intense pressure to evaluate novel information and make fast-paced decisions in a highly dynamic context. We show that successful open innovation projects are often shaped by immense productive bursts of activity set in motion by highly motivated volunteers. Case studies have shown that information overload during activity bursts impair management's ability to screen ideas and slow down decision- making. While productive bursts are an opportunity to harness the best of open innovation, managers should carefully design organizations able to cope with the rapid increase in activity levels. We offer six design principles to help managers deal with this challenge: 1) transparency; 2) self-censored clans; 3) emergent technology; 4) problem front-loading; 5) distributed screening; and 6) modularity. 2

16 Social Identification and Entrepreneurial Action: An Investigation into Hacker-Maker Communities Paper Maria Halbinger (Baruch College, Zicklin School of Business, CUNY) Francesca Melillo (KU Leuven) We study how an individual s relation with a social group regulates the choice to act upon an entrepreneurial opportunity that emerges from activities within the group boundaries. Drawing on social identity theory, we develop a theoretical framework to propose that high levels of social identification with a distinct social group inhibits entrepreneurial action and results in unexploited opportunities. However, this effect can be mitigated if other group members are engaged in the opportunity evaluation process. The analysis of survey data from 669 individuals in hacker-maker communities across the globe offers strong support for our hypotheses. This study highlights the importance of social identification for our understanding of entrepreneurial action, an aspect that both entrepreneurship theory and practice have so far neglected. 3

17 Triggers of Collaborative Prototyping: A Netnographic Study of User Innovation in Open Source Hardware Communities Paper Matti Grosse (Technical University Berlin) Volatilities in energy supply based on sustainable sources, as well as the need to adapt energy consumption levels in general, increases the demand for smart energy technologies. Although, a growing number of companies are engaged in providing smart energy solutions, users often experience them as not suitable for their specific purposes. This has spawned user-driven smart energy communities, which are developing open source solutions that can be easily applied and modified to local consumer needs. Following a netnographic research design that includes a comprehensive content analysis step, we analyze the OpenEnergyMonitor (OEM) community, one of the largest and most relevant open source hardware communities. Codifying 2,974 threads containing 20,407 posts, generated by 1,464 unique users resulted in a rich dataset that allowed us to empirically assess the determinants of collaborative prototyping in user innovation communities. We identify six distinct intentions or triggers that drive users to engage in online innovation communities: (1) help request, (2) asking for feedback, (3) providing feedback, (4) sharing information, (5) sharing product developments, and (6) calling for action. In a second step, we examined the impact of the identified triggers on the probability of collaborative prototyping and its successful completion by applying logit regressions. To enrich our analysis, we include several control variables on the characteristics of the thread starters as well as the users engaged in the respective innovation efforts. Our study helps to understand the triggers of collaborative innovation activities in user communities and to identify determinants of success of those collaborative user efforts. The results of our first model show that the provision of feedback and to call upon other users to innovate based on the OEM system, raises the probability of collaborative prototyping activities the most. Interestingly, we found that posts from users who ask other community members for their feedback on private innovation 4

18 efforts also significantly increases the likelihood to trigger subsequent collaborative prototyping activities. Further, we identified critical thread and user characteristics that are significantly positive predictors of collaborative prototyping activities and thereby confirming prior work on this topic. With respect to our second model, we identify that the provision of feedback and the request for feedback have the highest impact on the probability of a successful completion of collaborative prototyping activities in from of an innovation. Thus, feedback in either way plays a crucial role for the communities innovation efforts as well as its success. Nevertheless, the best predictor for success of a collaborative prototyping activity in a user innovation community is the level of innovativeness of the users participating in the respective process. 5

19 Who Moves to the Centre of Gravity? Developers Progression from Periphery to Core in the Linux Kernel Development Project Paper Marvin Hanisch (University of Passau), Stefan Berreiter (University of Passau), Carolin Häussler (University of Passau), Sven Apel (University of Passau) To coordinate activities between developers, open source software (OSS) communities assign roles of lateral authority to some individuals. Prior research has mainly described the characteristics of and functional roles within these virtual organizations, but few have examined which community members take on roles of lateral authority, or investigated what drives this progression of community members from peripheral roles to core roles. We develop and test a theoretical framework, which explains developers progression through technical contributions and communication activity. Using longitudinal data from the Linux kernel development project, our results suggest that both these factors of community engagement positively influence developers speed of progression. However, the relationships between community engagement and progression are weakened the more developers contribute to different subsystems and the more undirected their communication activity. In turn, developers responsiveness to the queries of other community members increases the positive effect of communication activity on the speed of progression. Our findings contribute to research on lateral governance in OSS communities, and to the more general debate on the benefits of specialization. 6

20 Open Radar Groups: The Integration of Online Communities into Open Foresight Processes Paper Michael Andreas Zeng (Helmut-Schmidt-University Hamburg), Hans Koller (Helmut-Schmidt-University Hamburg), Reimo Jahn (Helmut-Schmidt-University Hamburg) Monitoring a huge number of information sources and scanning the relevant environment are the basis for every foresight process. Since this endeavor is a complex task, it is necessary to find an efficient solution to do so. Based on foresight workshops with Radar Groups similar to Focus Groups, Delphi studies, and netnographies of online communities, we developed a pragmatic foresight process. Since we integrate online communities as information sources in our foresight process, we call this open foresight. This open foresight process focused the topic Aviation 2040 and was executed with aviation experts and an online community in two steps. This research setting enabled us to show (1) differences and similarities in future assessments and foresight capabilities between experts and online communities and (2) how to efficiently integrate online communities into foresight processes. Combining insights from experts and community members and, in doing so, moving towards Open Radar Groups seems to be a beneficial way for conducting foresight and enriching companies knowledge base. 7

21 Three Triggers of Search Why Young Firms Change Selective Revealing over Time Paper Joachim Henkel (TUM), Dominik Hepp (TUM) Capturing the right balance between benefits and risks from openness is a main challenge for organizations engaging in selective revealing. In order to find or sustain an optimal position, organizations change their level of openness over time. Taking a dynamic stance on openness, we propose a distinction between three types of such change triggers: unsatisfactory outcomes of earlier search processes, a change in conditions that renders a hitherto successful solution suboptimal, and a shift in what the firm defines as performance. Empirically, we draw on a multi method study linking qualitative and quantitative data from three segments of the software industry. By examining young firms that changed their engagement in selective revealing toward higher or lower levels, we find evidence of all three triggers. Specifically, conditions may change due to growth of the firm and an increasing risk of imitation, and the understanding of performance may shift from surviving and being a good community player to building a scalable business. 8

22 Legitimacy Creation in Grassroots Innovations: Community Supported Agriculture in the Netherlands Poster Ellen H.M. Moors (Utrecht University), Laura M. van Oers (Utrecht University) Global societal challenges emphasise the need for more sustainable modes of production and consumption in various sectors. Current strategies tackling these challenges reflect the dominance of market-driven technological innovations. This top-down approach designates a key role for firms in the innovation process and regards citizens as passive agents. More recently, civil society-led bottom-up grassroots initiatives are recognised as a significant societal movement, with the potential to shape transition pathways towards sustainability. Such grassroots responses are especially apparent in those sectors where problems are intensifying, existing models are failing and new possibilities are not adequately exploited (Seyfang et al.,2007, 2013; Vries et al., 2016). The agro-food sector is familiar with high grassroots activity, since sustainability issues have become increasingly important (Kirwan et al., 2013; White and Stirling, 2012). Farmers have seen their added value being captured by retailers and are looking for alternative ways of surviving. Conversely, consumers have become estranged from farming and have increasing demands regarding the quality, traceability and environmental friendliness of food products and processes. To challenge industrialised food systems, farmers and consumers have united in local food networks based on alternative values, principles, business models and organisational patterns. In particular, these initiatives set-up communities around growing and consumption of food and endeavour to re-connect consumers and producers; and re-localize agricultural production. Grassroots food communities include solidarity buying groups of local food; community supported agriculture and collective urban gardening initiatives. Grassroots initiatives provide alternative spaces to dominant practices within wider unsustainable regimes (Hargreaves et al., 2013; Seyfang and Longhurst, 2016). As such, grassroots innovations often face a mismatch regarding existing systems of norms, values and beliefs. Consequently, founders of such unconventional 9

23 activities need to cope with liability of newness. In particular, innovating entrepreneurs need to engage in efforts to get the innovation accepted as a desirable, realistic and appropriate i.e. legitimate alternative to incumbent substitutes in order for resources to be mobilised, for demand to form and to acquire regulatory support (Aldrich and Fiol, 1994). This paper focuses on legitimacy creation in grassroots innovation. Grassroots initiatives will have to constitute itself and attract members; they will have to raise funds and secure permission to operate. Throughout, they have to safeguard commitment and solidarity of their members. Principally, grassroots communities need to cultivate support and legitimacy locally. However, if they wish to endure and be influential, the initiative will need to seek approval from wider society. As grassroots innovation usually stems from the knowledge and experience of actors outside the formal institutions responsible for innovation, being taken seriously is a fundamental issue when trying to create momentum (Smith et al., 2015). This paper aims to understand the role of grassroots actors and strategic action to get the innovation accepted as a realistic and desirable alternative to incumbent substitutes (Bergek et al., 2008). The theoretical roots of this contribution draw on legitimacy types and legitimation strategy distinctions made within organisational sociology (Suchman, 1995). Specifically the paper focuses on the types of legitimacy and the legitimation strategies used to acquire these legitimacy types. The research design is an exploratory case study of Community Supported Agriculture (CSA) in the Netherlands. Around 46 Dutch CSA grassroots initiatives are known (Urgenci, 2016). CSA refers to a partnership between one or more farmers and a community of members (Flora et al., 2012). Dutch CSA presents a bottom-up induced niche in the highly industrialised Dutch agro-food system associated with numerous regulations and consolidation of economic power. Hence, it is expected that grassroots actors in the CSA niche need to engage in legitimacy creation to convince potential members, policy makers, the broader industry and special interest group of the desirability and necessity of CSA. Legitimacy creation in Dutch CSAs is explored by means of 25 interviews, and comparing them via synthesising patterns in legitimacy creation of the individual CSAs. This paper provides insights into a novel system of food provisioning based on bottom-up, community-led innovation processes. While innovation studies has extensively studied legitimacy at the level of technological innovations and industries, studies into the dynamics of legitimacy 10

24 creation of grassroots innovations are scarce and expected to enhance theoretical understandings on grassroots survival. 11

25 Delegation in Knowledge Creating Online Communities Poster Shiko Ben-Menahem (ETH Zurich), Yash Raj Shrestha (ETH Zurich), Georg von Krogh (ETH Zurich) Delegation involves a delegator s discretionary decision to defer authority and responsibility to a delegate (Yukl, 2013). Classical organization theory posits that delegation follows from a fundamental organizational problem of the division of labor, and correspondingly, the process through which organizational actors divide overarching organizational goals into subgoals (e.g., Puranam et al., 2014). Delegating authority and responsibility over subgoals to subordinates enables delegators (e.g., managers) to economize on scarce time and attention, utilize dispersed expertise more efficiently (e.g., Dobrajska et al., 2015), and motivate employees (Gambardella, Khashabi, & Panico, 2014). While offering insights into the rationale, process, and outcomes of delegation (e.g., Dobrajska et al., 2015; Leana, 1986; Yukl & Fu, 1999), prior research has been based on two questionable assumptions. The first assumption is that delegators can process information on potential delegates prior to choosing whether to delegate and to whom to delegate. For example, studies have shown that delegators rely on a formal reporting structure and role descriptions, congruence between the delegator and delegate s goals, and knowledge of the delegate s job competence, trustworthiness, and interest in assuming increased responsibility (e.g., Leana, 1986, 1987). The second assumption is that delegators operate within a hierarchy where they can use mechanisms such as formal control, incentives, knowledge transfer, and integration tools to minimize the uncertainty underlying the decision to delegate (e.g., Dobrajska et al., 2015; Gambardella, Panico, & Valentini, 2015). These assumptions are contested by recent work on new forms of organizing, which challenge the traditional mechanisms enabled by hierarchy and information on the identity and capabilities of the delegates. We aim to increase understanding of delegation in new forms of organizing to shed light on the microfoundations (Baer, Dirks, & Nickerson, 2013) of organizational delegation i.e., organizational actors decisions and behaviors relating to maintaining or delegating responsibility for 12

26 executing tasks. We do so by focusing on an organizational setting where (1) information on potential delegates identity and capabilities is severely limited, and (2) delegators lack the formal authority, based on hierarchical structures, to impose tasks on delegates and control their execution. Given the pervasiveness of new forms of organizing where delegators lack information and authority, we propose that current knowledge on delegation must be complemented with a new theoretical frame wherein the assumptions of hierarchy and information on delegates are relaxed. We conceptualize this as peer-to-peer (P2P) delegation and argue that in the absence of traditional design attributes, organizational members base their decision to either self-implement or resort to P2P delegation on the basis of (1) their prior experience with coordinating idea implementation (i.e., successful or failed self-implementation versus delegation) and (2) the complexity of the task at hand. Drawing on organizational learning theory concerned with the distinction between learning from success versus failure and experiential versus vicarious learning modes (e.g., Deichmann & Ende, 2014; KC, Staats, & Gino, 2013; Madsen & Desai, 2010), we disaggregate delegation patterns and ask the following research question: How does a decision maker s experience from coordinating prior implementations and task complexity impact delegation decisions and their subsequent performance outcomes? We will explore this question in the context of OpenStack an online OSS community with 34,024 independent and firm-sponsored contributors from 557 member firms spanning 177 countries. OpenStack has a technology-enabled organizational design in which community members take part in collaborative knowledge creation, problem solving, and decision-making. To balance learning from peers and strengthening reputation by submitting high-quality contributions, community members need to decide when to self-implement their novel software development ideas and when to delegate responsibility for implementing these ideas. We will employ the OSS setting to study how this decision is impacted by the mode (i.e., self-implemented vs. delegated) and outcome (i.e., failed vs. successfully implemented) of prior experiences. OSS development communities are an opportune research setting for studying delegation and individual learning for at least two reasons. First, delegation is particularly challenging in OSS communities because of the likelihood of delegating tasks to individuals without prerequisite expertise. In online communities, delegators generate ideas for software development yet delegate 13

27 the implementation. Delegators have little discretion over who implements the task once they defer their responsibility for implementation. Rather, delegates self-select into the implementation of these open tasks. Second, because learning-by-doing is particularly potent in OSS participation (e.g., von Krogh et al,, 2003), OSS communities are suitable environments for developers to enhance their learning (Hars & Ou, 2002; Hertel, Niedner, & Herrmann, 2003; Lakhani & Wolf, 2005). Initial data collection of about 9,000 software development ideas has been completed. All data on the delegation structure and implementation status of each software development idea, and all data on the prior experience of each developer, have been collected, and about 60% of the data on developers prior network ties, and 70% of data on task complexity have been mapped. We use logistic regression to estimate the likelihood that a given task was delegated and that it was successfully implemented (Deichmann & Ende, 2014; KC et al., 2013; Madsen & Desai, 2010). We will apply a mediation analysis using a structural equation modeling framework (Muthén, 2011) to explore the impact of task characteristics and developers prior experience on the likelihood of successful idea implementation through delegation. 14

28 Combining Crowdsourcing and Toolkits to a Business Model: The Case of Stata Corp. Poster Kathrin Reinsberger (Vienna University of Economics & Business), Nikolaus Franke (Vienna University of Economics & Business) Users increasingly participate in both defining and co-creating value via user toolkits (Franke & Pillar, 2004). To date, research has mostly focused on the cocreation experience (Kohler et al., 2011; Nambisan & Nambisan, 2008; Prahalad & Ramaswamy, 2004) and the abilities of customers that qualify them for participation in new product development via toolkits (Franke et al., 2006; Schreier et al., 2007). However, as co-creation is an emerging concept, it does not have a single dominant business model at present. One of the most important aspects of a successful business model is that firms address the needs of their current (and future) customers (e.g. Chesbrough, 2007; Teece, 2010; Zott & Amit 2011). There are countless examples of co-creation business models, including LEGO, Linux, and Wikipedia. One particular successful and noteworthy corresponds to the Stata Corporation, which hosts with its statistical software a long-term success in managing a co-created platform product. The main characteristic of Stata s co-creation business model is based on the symbiotic interaction between the user-community (Statalist) and the user-toolkit (Stata software platform). Based on the increasing role of crowdsourcing as a way of opening up a firm s business model towards external partners (e.g. Bogers, Afuah & Bastian 2010; Boudreau & Lakhani, 2013), this study contributes to complement the view on co-creation by focusing on the combination of crowdsourcing with a user-toolkit in particularly exploring the business model behind it. The research addresses the following research questions: 1. How does the business model of Stata Corp. work concerning e.g. its structure, characteristics, performance, benefits, etc.? 2. What are the incentives and motives of innovative users to freely reveal their software code to Stata Corp.? 3. What are common success factors for co-creation in business model innovation that can be derived from this? Stata Corp. is a US based software company, 15

29 founded in 1985 and since then producing and distributing only one single software product at present version 14 released. The product offered by Stata Corp. presents a software package for performing complex statistical analyses. The package offers the functions of a toolkit for user innovation and Stata encourages its customers to create and share new software codes for executing novel statistical techniques. The company then selects user developments of interest to many users, and adapts and incorporates these into its next product release. The success of the business model is obvious - 82% of the available commands build on user developments (Stata Corp., 2017). One additional reason to select Stata Corp. as our case study was its distinctive degree of openness regarding its business model. While SPSS represents a closed model users do not have the freedom of designing individual applications - R is developed by users and thus characterized as a full open source model. The business model of Stata Corp. instead is called a mixed or hybrid model - based on an incremental operating system with a small kernel that is closed, users are equal to Stata developers with respect to create add-ons to the software. In order to gather a complete understanding of the phenomenon in its context, we chose a case study design (Yin, 2014) based on a triangulation of methods. First, we intend to identify relevant information via an extensive secondary data analysis. In a second step, we aim at collecting primary data via semi-structured interviews with experts and managers of Stata Corp, to reflect the firm s perspective with regard to the research questions. Third, based on the data gathered from these two sources and existing literature, we want to develop a large-scale survey, which will be distributed to the around 2,500 users subscribed at Statalist community. The quantitative study yields at uncovering the user s perceptions and intentions to, for example, freely reveal their novel statistical techniques. Expected contribution We aim to enrich the open and user innovation research in the following: We provide valuable insights into a successful example of a business model, based on the interaction between a user-toolkits and a user-community (Franke & Pillar, 2004; Franke et al., 2008; von Hippl, 2001). This will not only help to further enhance the design and usability of user-toolkits but also illustrates exemplarily the implementation on when (reasonability) and how (basic essentials) to collaborate best with a usercommunity using a toolkit (Fueller, 2010; Shah, 2006). By focusing on the motives and incentives of users to freely reveal their developments, this study 16

30 contributes to a better understanding of the underlying factors of revealing and sharing intellectual property by using user-toolkits (von Hippel & von Krogh, 2006; von Hippel, 2005; Roberts et al., 2006). At present, this study is in its conceptualization phase. By July 1st 2017 we expect to have made significant progress regarding the study design and the operationalization. 17

31 Teaching User Innovation through Competition The challenge of cross-functional teams in digital learning environment Poster Minna-Maarit Jaskari (University of Vaasa), Mona Enell-Nilsson (University of Vaasa), Jussi Kantola (University of Vaasa) The role of user innovation continues to grow as more people become familiar with the concept. Higher education in universities of technology and multidisciplinary universities is a logical way to raise the awareness of the concept both in academia and outside academia. This poster presents a new research project focusing on student learning experiences and observations made by the participants (both students and teachers) in a user innovation teaching context at the University of Vaasa, Finland. A pilot user innovation course including multiple challenges was carried out: Teaching user innovation through competition between cross-functional teams in a digital learning environment. The aim of our poster is to describe the structure of the pilot course and to reflect upon student learning experiences and participants observations regarding the multiple challenges. One of the challenges in teaching user innovation is to balance between how to give some structure to students in order to enhance the learning process, but so as not to either hinder their creative thinking or thwart new innovative solutions. One central element of the piloted course was the cross-functional cooperation in the teams, i.e. people from different disciplines, ranging from technology disciplines to business and communication studies, working together towards the same goal. Crossfunctional cooperation has become an established strategy in new product development. In particular, companies seek to bolster cooperation between marketing, design, and engineering even though they may also combine other functions such as manufacturing. Different functions enable a project team to gather more diverse information and share ideas about what is commercially important, technically feasible, or difficult to manufacture. However, even if the literature supports cross-functional integration in the innovation process, companies often struggle to manage this integration successfully. Thus, working 18

32 in cross-functional teams during education can be an essential and fruitful learning experience. We therefore follow the argument that students should have a taste of cross-functional teamwork already during their studies as higher education provides a safe environment for a trial and error process. However, we note that if the cross-functional team members have different backgrounds, this may increase the probability and level of conflict within the teams. In the piloted course the different cultural background of the students also affected how they experienced the competition element; the pilot course was given in English and the group was international with more and less competition oriented students. Earlier studies suggest that successful projects are not distinguished by the absence of barriers or conflicts but rather by how the members overcome them. In poster presentation we identify different ways the students used to overcome such barriers. Furthermore, our poster raises the challenge of a digital learning environment for fostering team innovation, suggesting hybrid learning as a possible form for higher education. 19

33 Innovating beyond Firm Boundaries: Effects of Resource Deployment Control in Open Source Software Development Poster Mario Schaarschmidt (University of Koblenz-Landau) Many software-producing companies have started to make use of freely available software components that complement their own software module portfolio. Such freely available software is usually published under an open source software (OSS) licence (Santos et al. 2013). To better align development activities that take place in a community that surrounds OSS projects with own development activities, software firms have started to send own developers to work for these projects (Spaeth et al. 2016). For example, nine out of the 20 core developers for the Linux kernel work for RedHat, a OSS software distributor. However, a question that remains fully unaddressed in the discussion of open source software development is how firms exactly try to align what is developed in-house with what is produced inside the community (Schaarschmidt et al. 2015; Teigland et al. 2014). Schaarschmidt et al. (2015) introduced a framework that helps balancing control approaches that are designed to influence innovation activities that take place beyond firm boundaries. In particular, they distinguish a control-by-leadership approach, which related to pursuing leadership positions within a community, from resource deployment control. Here, firms send own paid developers, socialized within the boundaries of the firm, to work in an OSS project. The underlying rationale is quite simple: Developers that are socialized with the firm and which can be controlled via classical contracts work for the OSS project and try to align development trajectories with firm strategies. And the more developers a firm sends, the greater is the influence a firm can gain over the project. Until now, the concept of resource deployment control has not received much attention in the academic literature. Therefore, the aim of this research is to unravel determinants of resource deployment control, along with potential outcomes. In particular, if the concept as such is applied, a developer that is forced to work in an OSS project is likely to display opinion leadership (Iyengar et al. 2011). Moreover, this relation should be more prevalent for firms 20

34 that run a business model that is dependent on the OSS project. We decided to investigate the research question by means of a survey. First, we started our research by searching for OSS projects that have a vital activity of more than one firm. We found Linux to be a case in point. Then, we extracted developer addresses from an online data base. The parser we used collected more than addresses for the time from Out of these addresses, we selected the addresses that where most active and invited a subsample of developers to our survey. Of these addresses, about 10% were unavailable. Finally, about 800 developers answered the survey of which 321 worked in a software company. We measured our constructs with established measures (e.g. opinion leadership), except for resource deployment control. As no available scale to measure resource deployment control exists, we used an index-based measure of what developers are forced to do. In particular, developers had to indicate whether or not they actively try to align development activities in the community with firm development or whether or not they influence discussions. A index of 8 (meaning that all eight answer options where listed as yes ) indicated high resource deployment control while 0 indicates no such control. We also asked whether or not the firm the developer worked for had a business model running on Linux. As a corollary, we also tested whether or not the perceived reputation of the employer is a determinant (Cai, & Zhu 2016). Table 1 reports the regression results. In model one, where all developers are included, resource deployment control has a significant relation to opinion leadership. A more detailed picture is shown in models 2 and 3. Here, the proposed positive effect is only visible in the group that indicated to have OSS business models on Linux. Finally, the interaction of resource deployment control and perceived external reputation is only significant for the business model group. 21

35 Open Source Software Developer-Community Role Conflict and Organizational Turnover Poster Mario Schaarschmidt (University of Koblenz-Landau) Open Source Software (OSS) is usually developed by a heterogeneous group of hobbyists, professional contributors and firms (Feller and Fitzgerald 2002, Grand et al. 2004). However, while much research attention has been devoted to the group of volunteers and their motivation to participate (e.g., Bitzer and Geishecker 2010), or to the role of firms in OSS development (e.g., Teigland et al. 2014, Schaarschmidt et al. 2015), we lack an understanding of how employed developers behave when they work for an OSS project. In particular, employed developers may face role conflicts stemming from different social norms and beliefs inherent in both organizational and OSS cultures (Rizzo et al. 1970). For example, there might be situations where firm interests do not map with what the community finds appropriate in terms of development trajectory (O Mahony and Bechky 2008). If customer request functionalities that contrast community development, employees are pushed into situations where they have to balance firm interests and community interests (Daniel et al. 2011). Several authors have highlighted that employees find themselves in situations of dual allegiance (see Chan and Husted 2010), a state which may have detrimental effects (George and Chattopadhyay 2005). This research addresses the question if role conflicts exist as a consequence of dual allegiance and whether they are associated with increased turnover intentions toward employing organization or not. For this study, a survey design was chosen to answer the question. I surveyed 185 employed OSS developers and found that (1) ideology imbalance drive role conflicts and (2) that role conflicts are indeed related to increased turnover intentions with regard to the employing organization. No such effect was found for the association between role conflict and community (i.e. OSS project) turnover intention. In addition, I tested several moderators of the significant relationship. Here, perceived supervisor support dampens the positive effect of role conflict on organizational turnover intention. The results yield important 22

36 implications for management and theory. First, the research sheds light on the notion that there are role conflicts for employed OSS developers that may have detrimental effects. Second, this research shows that role conflicts lead to increased turnover intentions towards the employing organization but not for the community. Managers must be aware of this fact as employees who feel role conflicts are likely to quit and may be hired by competitors. However, this study also shows that managerial support for OSS development may be an important mechanism to overcome the conflict-quitting intention circle. To assess employed developers perception of role conflicts and turnover intention, I designed a questionnaire that comprised 47 question related to several wellknown constructs such as organizational identification, role conflict, ideology, job satisfaction, and turnover intention. For all constructs, multi item measures were used that were anchored on a 7-point-Lickert-scale ranging from 1 = fully disagree to 7 = fully agree. The questionnaire was administered via several discussion forums for professional open source development. Some developers then took the questionnaire and posted it on their private twitter account. Overall, 1393 individuals started to answer the survey of which 185 completed the survey and indicated to be employed open source developers (Response rate of about 13.28%). As in other studies concerning open source, more than 95% of participants were male. All constructs exhibit acceptable reliability and convergent validity as indicated by Composite Reliability Values of about.74 and higher and an average variance extracted that is higher than.58 for all constructs. In addition, I ran a confirmatory factor analysis using SPSS AMOS 22. Results indicate that all items load on respective constructs, which is especially important as I used the same items in relation to the employer and the community. I also ran several tests for identifying common method variance, that is, the variance between independent and dependent variable is more due to measurement than to actual theoretical connections. All tests (e.g. Harmans single factor test, common latent factor test) indicated that common method variance is not an issue. This research is still in progress and the results reported here may not be treated as final. Until now, I ran several OLS regressions with organizational turnover as the dependent variable using the SPSS macro Process. Thus far, results indicate that role conflict is indeed a driver of increased turnover intention (b=.30, p<.001). I also found an interaction effect for supervisor support. Here, employees who perceive little supervisor support are 23

37 more affected by role conflicts (in terms of quitting intentions) than those who perceive supervisor support for OSS development. Apart from limited conceptual and qualitative work (e.g. Chan and Hursted 2010), we observe a relative lack of understanding how dual allegiance (or loyalty) is draining employees. This research shows that role conflicts (perceived as a consequence of ideological misfit between firm and community) may be an important driver for turnover intentions. For managers, this implies to strive for a balance between own interests and community interests to prevent employers from feeling pressured to leave the company. Also supervisor support seems to be an important instrument to decrease employees turnover intention. This study is (even in its early stage) not free of limitations. First, I only use single source, self-rated data. As turnover intentions occur over time, a longitudinal or predictive study design would be more appropriate. Second, until now, only a limited set of controls has been considered (i.e. age and organizational tenure). There are a bunch of other factors one should control for such as job satisfaction and type of employment (full time vs. part time). 24

38 Contests, Crowdsourcing and Open Innovation 25

39 An Organization with many Faces: A Crowd-Based Venture s Versatility in Resource Mobilization Paper Thomas Gegenhuber (Johannes Kepler University Linz), Robert M. Bauer (Johannes Kepler University Linz) Under the umbrella of the cultural entrepreneurship literature, scholars seek to shed light on how ventures use cultural strategies (e.g. symbolic action, narratives) to aid resource-holding audiences in making sense of their venture and stimulate the provision of resources. With our study, we seek to fill two gaps in the literature, namely, a holistic understanding of resource mobilization strategies (i.e. material and symbolic practices) and going beyond dyadic situations (e.g. a venture addressing more than one stakeholder simultaneously). The empirical context for our investigation is the organizing practices of a crowd-based venture, which connects clients having innovation- and design problems with creatives (i.e. the crowd) providing solutions to these problems. We find that both material and symbolic practices can be adapted to or enhance the social world of a resource holding audience. Mechanisms, such as separating, coalescing and balancing allow for addressing multiple audiences simultaneously. 26

40 The Double Selection Environment: Ideas Selected by Crowds and Experts Paper Lars Frederiksen (Aarhus University), Michela Beretta (Aarhus University), Dirk Deichmann (RSM, Erasmus University Rotterdam) While experts constitute the traditional approach to evaluate and assess ideas for innovation, it is increasingly common for organizations also to involve crowds in decision-making. However, research is lacking on how both types of audiences select ideas in different ways. We examine which innovator and idea-related attributes are important to which audience in influencing their selection of ideas. Based on data collected from an online ideation platform over the course of 11 months, we show that crowds and experts decisionmaking processes are driven by different mechanisms. We find that crowds, in contrast to experts, are more easily influenced in their decision-making by past success of an innovator and by how ideas are formulated (i.e., the complexity of the idea description). In contrast to crowds, we find that experts discriminate against novelty by favoring ideas that are similar in content to other ideas. 27

41 From Few to Many: Scaling Crowdsourcing Design Platforms Paper Thomas Kohler (Hawaii Pacific University), Lea Rützler (Universität Innsbruck) Crowdsourcing has expanded from sourcing ideas to generate a wide range of outcomes. Crowdsourcing design platforms connect companies with design needs and creatives who complete design tasks. For companies, these increasingly visible design platforms provide a compelling alternative to insourcing or outsourcing design. By leveraging designers as external innovators, the platforms themselves can grow significantly in size and revenue without equally increasing its costs. However, the crowdsourcing design landscape is competitive and scaling platforms is challenging. We examined the leading crowdsourcing design ventures to identify scaling challenges and strategies. This article presents scaling paths and strategies for managers of crowdsourcing ventures as well as executives of traditional companies, who seek to build scalable crowdsourcing platforms. 28

42 Too Many Cooks Spoil the Broth: Community Feedback and Idea Quality in Idea Contests Paper Isabella Seeber (Universität Innsbruck), Daniel Zantedeschi (University of South Florida), Anol Bhattacherjee (University of South Florida), Johann Füller (Universität Innsbruck) Idea contests represent a novel and popular approach for organizations to leverage the creativity of the crowd for product and service innovations. In this approach, ideators present their initial ideas to a community of potential users and solicit their feedback for idea improvement or refinement. However, it is not clear under what conditions feedback lead to the development of better ideas. In this study, we examine the role of community feedback on idea quality an online idea contests by integrating feedback intervention theory with social capital theory. While doing so, we develop a set of hypotheses relating community feedback and idea quality and then testing those hypotheses using data from ZEISS VR ONE idea contest. Our empirical analysis suggests that more feedback on very similar ideas has detrimental effects on idea quality. When investigating feedback in more detail we can show that in principle task information feedback does lead to improvement in idea quality, while task learning and task motivation feedback does not. However, the number of users providing feedback plays an important moderating role. We can show that increases in users and task learning feedback affects idea quality negatively, while increases in users and task information have positive effects. Implications of our findings for theory and practice are discussed. 29

43 How Robust are the Results? A Bayesian Averaging Approach for Tackling Replication and Model Uncertainty in Research on Inbound Open Innovation Paper Bernd Ebersberger (Management Center Innsbruck), Fabrice Galia (Burgundy School of Business), Keld Laursen (Copenhagen Business School), Ammon Salter (University of Bath) In this paper, we explore the effects of inbound open innovation on firm s innovation performance. Empirical strategic management research in general, and research on open innovation, in particular, is subject to an important degree of model uncertainty. This is because the true model, and therefore the selection of appropriate explanatory variables, is essentially unknown. Drawing on the literature on the determinants of innovation, and by analyzing innovation survey data for France, Germany, and the UK, we conduct a large-scale replication using the Bayesian averaging approach of classical estimators. We test a wide range of determinants of innovation performance suggested in the prior open innovation literature, and establish a robust set of findings on the variables which shape innovation performance. We provide some implications for innovation research and explore the potential application of our approach to other domains of research in strategic management. 30

44 Entrepreneurial Ventures as Sources for Innovation? How Incumbents (Under)Utilize Corporate Incubators and Accelerators as an Open Innovation Approach Paper Sandra-Luisa Moschner (Hamburg University of Technology), Cornelius Herstatt (Hamburg University of Technology) Prior research has shown that investing into startups is a sufficient tool for interorganizational learning, harvesting innovation and engaging in entrepreneurial activities. Recently, a new model for corporate engagement with young ventures, which not involves corporate ownership, has gained popularity. In these corporate incubator or accelerator models both parties collaboratively advance venture creation during a short or medium-term period of time. First studies have analyzed that incumbents objectives for initiating such an engagement model are miscellaneous. Additionally, scholars have identified that the initiatives generally resemble an open innovation platform that allows corporate organizations to tap into knowledge from startups as external knowledge sources. However, content and mechanisms of knowledge flows from startups to incumbents are unclear. We analyze how established firms assess the value of new firms as external knowledge sources within the context of corporate incubators and accelerators. Further, we explore how incumbents obtain knowledge from entrepreneurial ventures during the startup support process. In order to shed light onto this new phenomenon, we use interview data from 13 corporate startup support programs (21 interviews) from various industries in Germany. Our findings show that established companies value young ventures as an important external knowledge source for innovation. The majority of firms is not only interested in technical (solution) knowledge or ideas related to their core business, but also in procedural and methodological knowhow. Strikingly, we find that only few programs contribute to open innovation collaboration during the three-stage support process or subsequent formal partnership between established and new firms. Consequently, most incumbents of our sample underutilize corporate incubator and accelerator programs as an 31

45 open innovation approach. Our study contributes to the literature by identifying startups as external knowledge sources for corporate firms with regard to the explicit and tacit knowledge dimension. We also show that corporate incubators and accelerators have the ability to foster collaborative open innovation between both parties, but surprisingly often fail to do so due to suboptimal startup selection and program structures. 32

46 Mapping Open Innovation in SMEs and Large Companies in KSA: A Case in the Industrial City in Riyadh Paper Amal Al-Dulaigan (Arabian Gulf University), Afaf Bugawa (Arabian Gulf University), Odeh Al-Jayyousi (Arabian Gulf University) The industrial sector is a pioneer in adopting a number of innovations including product, process, technological, business model and organizational innovations. SMEs play a crucial role in harnessing open innovation which contributes to enhancing competitive advantage and productivity. This paper aims to map open innovation practices in SMEs and large companies operating in Riyadh Second Industrial City, Kingdom of Saudi Arabia (KSA). For this exploratory research, a survey was developed to gather data about the status, motives, practices, and perceived challenges and barriers for 66 SMEs and 26 large companies from different industries during the period between The analysis shows that there are no significant differences in implementing open innovation practices between SMEs and large companies nor in the adoption barriers. Besides, both SMEs and large companies use more inbound than outbound open innovation activities. The study recommends that SMEs and large companies should expand their choices among a range of different inbound and outbound open innovation practices. Moreover, in light of economic transformations and vision 2030 of Saudi Arabia, SMEs and large companies should develop a strategy for open innovation so as to leverage their core competencies. Keywords: Open Innovation practices, mapping open innovation, SMEs, Industrial sector, Kingdom of Saudi Arabia. 33

47 Attenuating the Not-Invented-Here Syndrome: The Value of Behavioral Remedies Paper Julian Hannen (RWTH Aachen University), David Antons (RWTH Aachen University), T. Oliver Salge (RWTH Aachen University), Frank T. Piller (RWTH Aachen University) This paper presents two studies on possible remedies against the Not-Invented- Here Syndrome (NIHS), a persistent decision-making error arising from an attitude-based bias against external knowledge. Study 1 draws on 32 interviews and three focus group meetings with R&D employees to identify NIHS remedies applied in actual practice. Study 2 draws on quantitative data from 583 global R&D projects and has a twofold purpose. First, it quantifies the NIHS effect and shows how it impedes external knowledge acquisition and undermines project performance. Second, it applies our proposed behavioral approach and examines two specific debiasing strategies (perspective taking and considering the opposite) as effective remedies that help attenuate the negative effects of NIHS. These behavioral strategies complement rather than replace established remedies to mitigate the negative consequences of the NIHS. 34

48 Enabling Community-based Crowdsourcing: The Role of Open Innovation Intermediaries and Co-creation Capabilities Paper Krithika Randhawa (University of Technology Sydney), Ralf Wilden (University of Technology Sydney) This paper investigates the capabilities open innovation (OI) intermediaries deploy to enable their clients in implementing crowdsourcing via online communities. We use an exploratory case study of an OI intermediary and its client organizations to develop a theoretical framework of the intermediaries capability portfolio in helping clients address internal challenges to communitybased crowdsourcing. We find that technological and marketing capabilities are crucial for intermediaries to assist clients in overcoming project team-level and organizational-level barriers respectively. Furthermore, intermediaries leverage co-creation capabilities as a higher-order capability to further reinforce their technological and marketing capabilities. Comprising a portfolio of product- and market-oriented service capabilities, co-creation capabilities are the key to deploying client-centric services through which intermediaries co-develop both the product and market for online community engagement. Thus, our study uncovers the complementary role of co-creation capabilities in supporting clients deal with project-level and organizational challenges. These insights can form the basis for managers of intermediaries in targeting how client engagement capabilities are deployed to facilitate community-based OI. 35

49 Crowdsourcing without Profit: How Seeker Motivation and Strategies Drive Local Community Innovations Paper Krithika Randhawa (University of Technology Sydney), Ralf Wilden (University of Technology Sydney), Joel West (Keck Graduate Institute) Research on crowdsourcing as an open innovation (OI) mechanism has predominantly focused on profit-seeking firms. In this paper, we investigate how crowdsourcing is deployed by non-profit seeking organizations, and how it differs from the more familiar corporate context. Based on a sample of local governments that use the same intermediary, we examine how varying seeker motivation and strategies drive differences in their online engagement behaviour, in turn shaping the likelihood of crowdsourcing success. We develop a three-phase model of crowdsourcing implementation showing how seekers with a strategic approach to crowdsourcing, driven by strong motivation for societal transformation, are more likely to yield successful results from their crowdsourcing initiatives. Our findings reveal that the non-pecuniary orientation of both seekers and solvers makes the motives, goals and processes of such crowdsourcing fundamentally different to corporate crowdsourcing. Further, the local pool of solvers in this context resembles a cooperative community rather than the competitive crowds prominent in for-profit crowdsourcing. Our insights have implications for OI efforts of both non-profit and corporate organizations. 36

50 Pressure in Crowdsourcing Contests Paper Jonas Heite (Max Planck Institute for Innovation and Competition), Karin Hoisl (University of Mannheim) Our paper contributes to the literature on performance in contests by providing new and causal evidence of the mechanisms causing performance differentials that cannot be explained by differences in ability. Existing work agrees that competitive pressure, i.e. competing against contestants with a higher ability, decreases the performance of medium-ability contestants. This outcome is explained by a reduction in effort if contestants think that there is only a small likelihood that they can win. This is, however, only one possible explanation for their decrease in performance. Status seeking, which might entail taking higher (or too high) risks, may be another explanation. To shed more light on the mechanisms causing these performance differentials, we use data on crowdsourcing contests hosted on the topcoder platform. The data allow us to implement a Regression Discontinuity Design analysis. We study individuals with the same abilities. Part of these individuals, however, compete as top-performers of a low-ability group, the others compete as the bottom-performers of a highability group. Hence, competitive pressure should be higher for the latter individuals. In a first step, we compare the performance of the individuals competing in the two groups. In a second step, we investigate the effect of star performers, i.e. contestants with a skill rating above the 95th percentile, on the performance of medium-ability contestants. Star performers should increase competitive pressure. Our results confirm that bottom-performers of a highability group are characterized by a performance that is 75 points lower than that of contestants who have the same skill level but compete as top-performers of a low-ability group. The performance differential between the two groups equals 36 %. Hence, the effect is not only statistically but also economically significant. The second analysis, restricting the contests of the high-ability group to those with star performers, shows that the difference between the high and the low-ability groups increases to 83 points (performance differential: 57 %). To investigate the mechanisms causing the above summarized results, we 37

51 investigate the behavior of the contestants. We analyze their choice of tasks (tasks vary in the level of difficulty) and their speed of problem solving. Based on various descriptive statistics of the problem-solving behavior of the contestants, we find that the bottom-performers of a high-ability group tend to try to solve more difficult problems than the top-performers of a low-ability group, potentially to signal their abilities. The difference becomes even more pronounced once star performers are among the contestants in the high-ability group. In sum, we find first evidence that status seeking might be another explanation for performance differentials under high competitive pressure. 38

52 Challenges and Capabilities to Profit from Customer Co-creation in Organizations Paper Christiane Rau (University of Applied Sciences Upper Austria), Katja Krämer (University of Applied Sciences Upper Austria), Kristiana Pavlova (University of Applied Sciences Upper Austria) Many firms are facing challenges by applying customer co-creation approaches. Whilst recent research provides extensive insights on the formal methods and theoretical concepts, there is still a research gap concerning the intraorganizational implementation of customer co-creation projects. This piece of work attempts to close this particular gap by exploring the organizational capabilities in customer co-creation projects based on an analysis of twenty interviews with innovation managers and intermediaries. As a result, we identify four recurrent challenges and related capabilities, i.e. (1) to set up a co-creation project for the first time, (2) to manage outcome expectations, (3) to evoke & keep customers interest, and (4) to integrate customer co-creation and NPD processes. Connected to each theme, we find that organizations need to develop critical interrelated structural, procedural, and communicational capabilities. Particular activities, we name them capability enhancing activities, can support the organizations in developing these capabilities. 39

53 Is Open Innovation in Startups DNA? Paper Zeljko Tekic (Skolkovo Institute of Science and Technology) In this paper we attempt to shed more light on close, but not completely understood relationship between open innovation and startups. Analyzing the process of startup development through the lenses of the open innovation principles proposed by Chesbrough (2003), we provide evidence that all six principles are clearly and intensively reflected in practices startups use to survive and grow. More than that, startups do not apply open innovation as an additional layer to the other practices, they crucially depend on it. Without using knowledge and expertise of people outside their organization, without experimenting with potential users and all other stakeholders, without relying on externally initiated R&D as an input for their own work, startups would hardly develop suitable business models in acceptable time horizon, and survive. Based on this, and accepting that the main differentiator between established firms and startups is not quantitative but qualitative one startups search for a business model, while established firms, being large, medium or small, execute established we challenge the existing understanding that the bigger the firm, the higher probability it applies and adopts open innovation. We argue that the level of adoption of open innovation practices is not principally determined by the size or age of a firm (quantitative characteristics of an organization) but by the stage at which an organization is in its search for a repeatable and scalable business model (qualitative characteristic of an organization). Organizations which search for a business model are more open than those which execute one. After the business model is validated, the quantitative characteristics dominate (more resources, more openness). 40

54 Unfolding Agile Innovation Capabilities in Co-Located Innovation Labs Paper Florian Fecher (University of Salzburg), Johanna Winding (Universität Innsbruck), Katja Hutter (University of Salzburg), Johann Füller (Universität Innsbruck) In today s digital age and dynamic market conditions a variety of agile and lean oriented formats such as innovation labs (ILs) evolved to conquer transformational and adjacent new business op-portunities. While the literature discusses the formats and their qualities to master innovation chal-lenges, little is known about the dynamics from both the organizational and individual perspec-tive. Our case study draws upon a unique dataset from an IL in the banking and financial service industry. The results reveal how to successfully set up and run ILs as well discuss the phase out, it s reintegration in the organization. Our findings show that above and beyond the physical space, resources and facilitator we have to attach a great importance to the participants expecta-tions and experiences. The results further our understanding of agile innovation capabilities and an ecosystem to accomplish end-to-end solution, the reintegration and realization of the ideas. 41

55 B2B Crowdsourcing for NPD and Innovation: A Literature Review and Research Directions Paper Sylvia Dimitrova (University of Padua) Nowadays, crowdsourcing is used mostly in the B2C industries, but its use in the B2B industries is also increasing. Nevertheless, the B2B use of crowdsourcing is a topic that has been underexplored to date and merit further investigation. In addition, the extant literature on crowdsourcing is spare with respect to synthesis of empirical research on the use of external crowdsourcing for NPD and innovation. Therefore, this paper reviews empirical studies on crowdsourcing for NPD and innovation, and on the use of crowdsourcing in B2B setting in particular. It summarizes results on how crowdsourcing is used for NPD and innovation and outlines benefits and challenges for its use. Additionally, the paper discusses the specifics of crowd integration in B2B crowdsourcing and summarizes benefits and challenges for the implementation of crowdsourcing in industrial context. Future research directions are outlined. 42

56 Consumers hijacked my idea contest! Managers preference formation and reactions towards deviant content in ideation contests Paper Alexandra Gatzweiler (Aachen University), Vera Blazevic (Aachen University), Frank Piller (Aachen University) In recent years, firms increasingly recognize consumers potential as an innovation partner and invite them to provide ideas or concepts addressing a challenge defined by the firm. While this open call for collaboration has been shown to deliver valuable input for the innovation process, it also provides an opportunity for unforeseen ideas and even the submission of illegal, unconventional and incongruous content, a behavior that represents deviant cocreation. Managers have to master a complex task when assessing the character of deviant content and are often unsure about the appropriate response and whether to implement the content. Given the increasing number of co-creation activities initiated by companies, this lack of proficiency and understanding can lead to potential negative outcomes for firms. This essay employs a grounded theory approach drawing on a set of interviews undertaken with 11 innovation managers, to explore managers preference formation and reaction towards deviant content as well as the influencing factors during these decision moments and the impact of deviant content in the contest and the firm. First of all, the study reveals that managers assess and react differently towards deviant content, providing evidence to the uncertainty that deviant content generates. The authors then conceptualize the decision process in ideation contests to better understand the decision moments and when a preference is build or the reaction towards the content is expressed. Furthermore, existing theories on decision-making in innovation are complemented to the disentangled process to show that deviant content provokes a shift of the managers decision making mode from automatic/intuitive to deliberate/complex and demonstrate how and when boundary conditions impact the individual decision-making process. The results help to provide a more balanced view on co-creation in innovation by 43

57 highlighting the arising challenges of managing ideation contests and leveraging consumers contributions within the firm and especially show that preferences and decisions about deviant content are highly context specific and dependent upon managers openness for unique contributions. 44

58 Organizing for Open Service Innovation: Moving Beyond Manufacturing Poster Deborah Roberts (Nottingham University Business School), Simona Spedale (Nottingham University Business School), Wolfgang Gruel (Stuttgart Media University) While open innovation (OI) is spreading across several industries there is a need to go beyond manufacturing contexts, where OI originated and where conceptual and practice-oriented developments have so far concentrated, to generate greater understanding of OI in services (Randhawa, et al., 2016; Chesbrough, 2011). Services form a significant part of the economy in developed countries and are expected to dominate projected economic growth in the 21st century (Papastathopoulou & Hultink, 2012). The need to redirect OI research from its manufacturing birthplace is justifiable on two grounds. First, pressure is escalating for manufacturing firms to embrace servitization (Lightfoot et al., 2013; Vandermerwe & Rada,1988) and differentiate through explicit service innovation strategies (Froehle & Roth, 2007; Kahn et al., 2006). Manufacturing firms that servitize can benefit from extending OI principles and practices to their ancillary services and by engaging in service open innovation (Chesbrough, 2011). Second, there is growing recognition that OI is critical for pure service firms (i.e. firms without a manufacturing core). These are, in fact, even more dependent on service innovation for successful differentiation and sustainable profitability than manufacturing firms as testified by the emergence of innovation in services as a distinct and growing research field (DeJong &Vermeulen, 2003; Menor & Roth, 2007). There is, however, little systematic understanding of the specificities of open innovation in services. Current knowledge is limited to generic characterizations of the differences between product and service innovation (Ettlie & Rosenthal, 2011) based on commonly acknowledged and recognizably distinctive features of services e.g. intangibility, inseparability, heterogeneity, and perishability (Zeithaml et al., 1985) or to very specific issues such as, for example, client-coproduction 45

59 (Mattsson, 2010; Menor & Roth, 2008). This project addresses the emergent interest in open service innovation by adopting a comparative approach. By using qualitative research that combines diverse data sources (including observations, documentary data and interviews), it compares in-depth OI case studies (Yin, 2009) in a large automotive company engaged in servitization and the cultural heritage industries. Here customer co-creation as a form of open innovation is at the core of service innovation. The choice of the sampled case studies is purposeful in that the project spans across several organizational dimensions that might be of interest for understanding open service innovation and its specificity and nuances more in depth. On the one hand, the large manufacturing company in the sample represents the traditional approach to open service innovation as an extension of OI principles to a different domain of activity within a long-established, permanent large organization in a sector the automotive industry which has been previously investigated; on the other hand, the heritage sector is represented by a temporary organizational unit (a small, independent multidisciplinary task-force) created ad hoc for the purpose of stimulating open service innovation in a mature sector. Recent developments in the heritage industry and, more particularly, the push to design exhibitions that provide more immersive, sensorial and affective visiting experiences in museums (Dudley, 2010), have paved the way for the adoption of co-design/cocreation approaches in projects that combine curatorial, technical and design competences (Petrelli et al., 2016). However, little systematic research has been carried out on open service innovation in this context. The project contributes to theory and practice development in two ways. First, the comparative approach allows for exploring potential differences and similarities between these two types of open service innovation, providing the basis for further theorization of open innovation in services and for the identification of managerial best practices to suit this context. Second, the project contributes to fleshing out the specificities of open service innovation as, potentially, distinct from open innovation in manufacturing, therefore contributing to ongoing scholarly debates. At the time of writing the first round of data collection has been completed and preliminary analysis is ongoing. The full results from the comparative analysis on the empirical data collected so far will be available by 1 July Early analytical efforts point to the significance of soft organizational dimensions in affecting the success of open service innovation 46

60 and, more particularly, aspects of culture (e.g. strength, type), time-frame and inertia, and leadership. 47

61 Gender Imbalance Online Poster Andrea Blasco (Harvard University), Karim R. Lakhani (Harvard Business School), Michael Menietti (Harvard University) Though the proportion of Internet users is nearly the same across the genders, researchers have found significant gender imbalances in online behavior (e.g., only 13 percent of Wikipedia contributors are women; less than 5 percent of StackOverflow active users are women). While online platforms do not seem to put in place discriminatory rules that can explain the gap, it seems that some institutional characteristics of the platforms might have inadvertently promoted an imbalance. In this study, we examine two possible mechanisms: i) the combination of gamification and incentives with differences in preferences between the genders; and ii) the impact that role models may have on online behavior. We use the results of a field experiment on the crowdsourcing platform HeroX to assess the effects of these mechanisms on gender imbalance online. 48

62 The Positive Side of the Corruption Effect: Inducing Controlled Motivation to Increase Accuracy in Crowdevaluation Poster Christian Garaus (Vienna University of Economics & Business) Organizations increasingly consider outsourcing evaluation tasks to the crowd. When ideas are abundant, crowdevaluation has been suggested to be superior to internal filtering in terms of efficiency and efficacy (Piezunka & Dahlander, 2015). Tapping into the wisdom of the crowd may seem all too tempting, if there were no other studies warning about the effect of individual biases, self-reinforcing processes, and strategic behavior that may turn the wisdom into madness (Mollick & Nanda, 2015). Prior research has demonstrated that at least parts of the madness can be explained by individuals who are motivated to distort results (Velamuri, Schneckenberg, Haller, & Moeslein, 2015). Research from related fields has suggested some reasons for doing so such as in- and outgroup discrimination (Tajfel, 1970), selfinterests (Dellarocas, 2003), and friendship bias (Love, 1981). When individuals are motivated by one of these reasons, biased estimates seem to be inevitable. The anonymity and lack of personal consequences that characterizes online voting seem to inhibit any mechanism that is targeted at reducing such motivationally based biases. A major question in crowdevaluation thus is how organizations can avoid such inaccuracies. Before considering this question in detail, it is important to understand a particular characteristic of this undesired behavior: The behavior is self-determined. In virtually every situation individuals have the autonomy to decide whether to vote according to their true assessment or if they provide a biased answer. Similar unwanted consequences of autonomous motivation have already been discussed by Montesquieu 250 years ago, but not been picked up by mainstream motivation research (cf. Osterloh & Frey, 2000). The solution put forward in Montesquieu s douxcommerce thesis is as easy as intriguing: autonomous motivation can be problematic, thus it should be replaced by behavior that is regulated through external instances. In the context of crowdevaluation, relying on controlled 49

63 motivation might indeed be the solution for getting unbiased estimates. But how could organizations induce such controlled motivation? Research on reward-motivation relationship has demonstrated that tangible rewards undermine autonomous forms of motivation (i.e. corruption effect; Deci, 1975). Particularly, if individuals receive their rewards based on the performance on a given task, the extrinsic motivation provided by the reward reliably crowds out any autonomous motivation previously held (see Deci, Koestner, & Ryan, 1999 for a meta-study) and individuals will try to maximize their payoff. As a result, strategic behavior and thus motivationally based biases should be less likely. Thus I hypothesize: In crowdevaluation contexts performance-contingent rewards will increase the accuracy of individuals evaluations. In order to test this hypothesis, I will first conduct a pilot study to identify how to best stimulate strategic behavior. A survey-based field study will be administered to individuals that participated in crowdevaluation in the past. Individuals will be asked, if and why they evaluated ideas better or worse than their true opinion. The most frequent answer will serve as stimulus for the experiment in the main study. The main study will employ a between-subject design with reason for biased evaluation and performance contingency as manipulated variables. Participants will be allocated to one of three groups: one control and two experimental groups. To create a realistic evaluation setting and thus to guarantee a certain degree of external validity, respondents will be exposed to ten fictions ideas and asked to evaluate them separately along the following dimension: novelty, nonobviousness, workability, relevance, and thoroughness (cf. MacCrimmon & Wagner, 1994). One of those ideas will present a statement that provides a reason for purposely biasing the evaluation (e.g., This idea was submitted by a friend of yours ). The idea containing the stimulus for inducing the intended bias will be selected randomly for every participant to avoid a potential biased based on personal preferences. The design will be identical for both experimental groups except for one difference: the presence of performancecontingent rewards in the controlled-motivation condition. In this condition, participants will be informed that they will receive a financial reward, if their evaluations come close to the true value, which is known for some of the ideas. In fact, the average of the control group will serve as baseline for assessing the accuracy of the individual estimation, as members of the control group have no incentive to distort their evaluations. The differences between the values of the 50

64 control group and the experimental groups will serve as dependent variable in subsequent regression analyses. 51

65 Self-Selection in Online Innovation Contests Poster Nikolaus Franke (Vienna University of Economics & Business), Philipp Topic (Vienna University of Economics & Business), Ines Reith (Vienna University of Economics & Business In recent years online innovation contests have gained popularity since the internet facilitates access to problem-solvers. In these contests, individuals decide freely, whether to join a contest (self-select in), or leave a contest (self-select out). Extant research assumes that this self-selection process is beneficial based on the premise that self-selecting individuals know more about their capabilities and knowledge than the publisher of the task (Geiger & Schader, 2014; Afuah & Tucci, 2012). Crowdsourcing research has mainly focused on investigating individuals who submitted their ideas (e.g. Brabham, 2010; Lakhani, Jeppesen, Lohse, & Panetta, 2007). In sharp contrast to its importance the principle of selfselection in crowdsourcing innovation contests is hardly investigated and hence understood. Almost all empirical studies on crowdsourcing innovation contests employ cross-sectional data sets consisting of the results of self-selection, i.e. a self-selected sample of participants. Little information is available about the counterfactual: individuals who self-select themselves out and the reasons for doing so. Innovation contests comprise a series of self-selection decisions and the set of surviving participants can both be a perfect self-selection or a loss of extremely potential contributors. There is no guarantee that the voluntary and open character of innovation contests leads to efficient self-selection of problem-solvers. Thus, our first research question is: (1) What are patterns of self-selection? - In a typical crowdsourcing contest: who selects in, who selects out? - When does this happen in the process? - What are critical stages? - In how far is there positive or negative self-selection? However, the questions is what drives self-selection. Our theoretical basis is motivation, which is portrayed as the main antecedent of self-selection in the crowdsourcing literature (e.g. Afuah & Tucci, 2012; Boudreau & Lakhani, 2011; Frey, Lüthje, & Haag, 2011). Motivation itself has to be differentiated in intrinsic (engaging in an activity because of interest and enjoyment of the activity per se) and extrinsic (engaging 52

66 in an activity because of external events such as rewards) motivational dimensions (Ryan & Deci, 2000). Based on motivation literature we hypothesize that they gain and loose importance over the time of the contest (e.g. Reeve, 2014, p. 20), but their added score always has to exceed a threshold value. Dropping below this value means a withdrawal (self-select out) from the contest, because the expected costs transgress the expected value. As a result, our second research question is: (2) What motivates participants to select in or select out? - Which dimensions of motivation (extrinsic/intrinsic, differentiated in its components) are critical for the self-selection decision? - Does the importance of different dimensions of motivation change over time and the different stages of a crowdsourcing contest? - Is the importance of different dimensions of motivation dependent on the incentive structure of the crowdsourcing contest? - Particularly interesting: The small population of wanted participants We want to investigate how dimensions of intrinsic and extrinsic motivation change over time and especially how the wanted population decides. Enhancing our knowledge on this important aspect is not only of academic interest, it also allows organizers to design their innovation contests more effectively. We will investigate self-selection in a unique longitudinal and real research setting: a panel provider will host an online innovation contest, posting a problem related to his innovation-activities (finding innovative ideas for a new app) to a heterogeneous population of 25,000 potential problem solvers. Based on interviews with crowdsourcing experts, we understand participant self-selection as a multi-stage decision process. Therefore, with the approval of the host, we will monitor each individual s self-selection status in the course of the contest using distinct online indicators, such as cookie tracking. After determining their self-selection status, we will send out surveys to random samples of individuals of the same status via the panel provider, asking individuals to participate in a survey as part of their panel membership. This unique approach will enable us to compare different self-selection groups (self-selected in or out) systematically in different points in time during the period of the contest, shortly after selfselection decisions were made. In addition, in order to determine the extent of self-selection in dependence of contest parameters, we will induce two different stimuli conditions (different incentives for participation) and monitor the selfselection response of individuals. Our goal is to finalize the data collection by July The expected results will, on one hand, decode the process of self- 53

67 selection, providing us with answers to the question which individuals self-select themselves in which direction (in or out) during an innovation contest. This will answer the important question whether self-selection actually works in innovation contests. On the other hand, the application of different contest stimuli will resolve the question of controllability of self-selection from a practical point of view. As a result, innovation contest organizers could enhance overall quality of their contests, if potential top-solvers are for example encouraged to self-select themselves into the contest and less-qualified participants prevented from participation. Novelty and importance of the project For the first time in crowdsourcing research, actual self-selection behavior of individuals in the course of an innovation contest will be observed in a longitudinal, real research setting which will enable us to interrelate individual-level data. We expect that this unique approach will allow us to make fundamental statements about the functioning, efficiency and consequences of participant self-selection in crowdsourcing systems. 54

68 Individual Search and Innovation Performance in Open Innovation Poster Anne Greul (Technical University of Munich), Tim Schweisfurth (Technical University of Munich), Christina Raasch (Technical University of Munich), Chia- Huei Wu (London School of Economics) Established companies need to explore new business solutions and serve future customer needs in order to avoid being disrupted by new entrants. As employees in organizations are the ones who actually search and identify new opportunities and create radical innovations with the potential to disrupt a firm s current core business, understanding individual-level search becomes crucial. While we know that search has a positive impact on innovative behavior, we do not know which factors initiate search and how individuals split their search effort between internal and external search. In this paper we aim to investigate what the antecedents and outcomes of individuals internal and external search are. We use a diary study approach to collect data on a weekly base over a period of twelve weeks. This research design enables us to measure within-person variance of employees idea generation. 55

69 Factors Influencing the Quality of Open Ideas: Open Innovation, Networks and Interlocking Ties Poster Nassim Belbaly (Montpellier Business School), Rajaa El Mezouaghi (Montpellier Business School), Calin Gurau (Montpellier Business School) Modern firms face competitive challenges that shorten the products life cycles and require a quicker rate of innovations. This necessity reveals a need for organizational change at research and development (R&D) level, often expressed through the implementation of open innovation strategies which connect the firm to external networks of knowledge and creativity. Although previous studies indicate that higher network connectivity leads to a better quality of outsourced ideas, little is known about the processes involved in building and maintaining strong interlocking ties between various network actors. In this paper, we investigate the importance of interlocking ties between various actors of the external innovation network, focusing mainly on the role of knowledge brokers as focal nodes of these networks and relationships. Adopting a grounded theory approach, our aim is to develop a model which provides a clear description of the factors enhancing the quality of ideas through the use of interlocking ties. Our results indicate two main categories of factors that may be responsible for the high quality level of the identified innovative ideas the actors and the processes involved in these interlocking ties. The results hold important implications for both the academic literature and the innovation management practice, by providing a deeper understanding of the levers of action that can be used to enhance the quality of the open innovative ideas. 56

70 Divide and Conquer: Experimental Analysis of Mechanisms that enable Crowd Filtering of Crowd-Generated Ideas Poster Victoria Banken (Universität Innsbruck), Isabella Seeber (Universität Innsbruck), Alexander B. Merz (Universität Innsbruck), Ronald Maier (Universität Innsbruck) Companies run innovation contests to publicly source new ideas, which may result in customer-focused business outcomes (Chesbrough, 2003; Du Plessis, 2007; Gassmann & Enkel, 2004; Nagar, Boer, & Garcia, 2016). In such contests, the crowd typically generates hundreds and sometimes thousands of potentially promising ideas (Bjelland & Wood, 2008; Jouret, 2009) that are subsequently filtered by domain experts. Until now, this filtering is time consuming and expensive as many experts are engaged in a complex decision making process (Nagar et al., 2016) to pick the few most original, unique, useful, and elaborated ideas (Dean, Hender, Rodgers, & Santanen, 2006). At IBM s Innovation Jam, for example, 50 senior executives spent one week to elicit the best 31 out of 45,000 ideas (Bjelland & Wood, 2008). To elicit 40 semi-finalists out of 1,200 ideas in Cisco s I-Prize competition, six full-time employees spent three months (Jouret, 2009). Those who filter such large quantities of ideas are not only faced with the challenge of an exceeding cognitive load imposed by this complex task (Sweller, 1988), but also by the issue of similar ideas occurring in substantial amounts (Kornish & Ulrich, 2011). Different approaches on how to improve idea-filtering processes exist. Some studies suggest to crowdsource certain decision-making activities formerly performed by experts in order to improve resource-efficiency (Burnap, Ren, Papalambros, Gonzalez, & Gerth, 2013; Grad & Vienna, 2016; Green & Seepersad, 2014). Scientific insights are scarce on how to subdivide the filtering process into activities that can be crowdsourced and activities that are better suited for experts, also with respect to the sequencing of activities. Moreover, other studies aim at introducing IT support to automate (parts of) the filtering process by utilizing text mining algorithms (Kruse, Schieber, Schoop, & Hilbert, 2013; Nagar et al., 2016; Walter & Back, 2013). We know little about how IT-enabled similarity identification of ideas can be embedded in this 57

71 filtering process as a form of automated pre-processing. Little research exists up to date that combines both approaches of a) crowdsourcing and b) IT-enabled similarity identification for idea filtering. This research intends to support contest hosts by investigating how to effectively involve the crowd in the filtering process. We do so by demonstrating effective sequencing of idea filtering activities and showing the benefit of IT-enabled similarity identification. Therefore, we will answer the following research question: How do IT-enabled similarity identification and the sequence of the filtering process affect the quality of the ideas selected? We plan to conduct an experiment using secondary data of an idea contest with 525 submissions. We want to adopt a 2x2 factorial design manipulating the process sequence and the screening support. The process sequence will be altered in that either categorization activities or screening activities come first. In categorization, experts define category labels and afterwards the crowd assigns all ideas to these categories. In screening, the crowd is asked to eliminate bad ideas. In the screening support, similar ideas will be presented together based on text-mining results assessing uniqueness in ideas. The assignment of categories and the screening of ideas is not tied to the same crowd workers. We will assess differences in processes by comparing the quality of final idea sets, required time and number of required individuals. Idea quality will be assessed by external raters, which had selected 43 ideas from the original 525 ideas as good ideas. We will measure number of ideas in bag after each activity as well as cognitive load, attention and motivation in order to control information overload and lack of concentration. Expected findings and implications We expect to contribute to theory and practice in different ways. As it should be possible to evaluate originality reliably with a large set of novice raters (Green & Seepersad, 2014), we will first show that the crowd can reliably eliminate similar ideas because we expect that not too much domain specific knowledge is required to screen out bad ideas. We will further show that screening out ideas pre-processed with text mining is associated with higher idea quality than screening out ideas without IT support. We assume cognitive load to be lower when the differences in the content of ideas within a subset of ideas are smaller. We thus expect that starting with the screening of the ideas will reveal more promising results than starting with the categorization of ideas. Therefore, experts will define categories grounded on a better information base that should not include similar ideas of which one or more are comparably bad. 58

72 The implications of our study will enable contest hosts on one hand, to more efficiently design the filtering process by enabling on the other hand, the crowd to reduce a high amount of generated ideas to its essentials with the help of IT. The research was partially funded by the Austrian Science Foundation (FWF): P GBL. 59

73 How Close are Open Innovation Partners? Degrees of Institutionalization of Firm-Consumer Collaboration for Innovation Poster Eva Lendowski (University of Münster), Ansgar Buschmann (University of Münster), Gerhard Schewe (University of Münster) How close are open innovation partners? Degrees of institutionalization of firmconsumer collaboration for innovation Eva Lendowski, University of Münster, School of Business and Economics & Dr. Ansgar Buschmann, University of Münster, School of Business and Economics & Prof. Dr. Gerhard Schewe, University of Münster, School of Business and Economics Motivation This research is part of a multi-stage research project investigating how consumer goods companies implement open innovation with consumers to better understand the boundary conditions of open innovation. Research so far has found extensive evidence of the benefits of open innovation with customers and users for organizations (Foss et al. 2011; Franke & Shah 2003; Mahr et al. 2014; von Hippel & Katz 2002; West & Bogers 2014). Consumers have been identified as the most valuable external contributor to new product success as products cocreated with consumers better fulfil market needs (Gemser & Perks 2015; Poetz & Schreier 2012). Collaborating with consumers for innovation has seen the biggest rise in consumer goods industries throughout the past decade (Lakhani & Panetta 2007; Simon et al. 2016). Due to the advent of digital media, innovating with a broad network of individual users and consumer communities has become more accessible for a wider range of companies. There are many ways to integrate users into all stages of new product development, from ideation to commercialization. Besides establishing consumer communities for exchanging knowledge and innovations with customers under their own brand name, companies can also cooperate with innovation intermediaries offering access to innovation platforms with large member networks (Bartl et al. 2012; Diener & Piller 2013; Fuchs & Schreier 2011; Hienerth et al. 2014; Piller et al. 2011; Rohrbeck et al. 2010). Additionally, innovating companies can make use of various Social Media channels, such as Facebook, to engage with consumers and 60

74 solicit feedback (Piller et al. 2012; Rathore et al. 2016; Roberts & Candi 2014). This preliminary study analyzes how customer co-creation is institutionalized by consumer companies. We compare the three most extensively used digital channels to integrate consumers into new product development: firm-hosted communities, communities managed by intermediary organizations, and social media. Our overarching aim is to advance theoretical and empirical understanding of the interaction of innovation objectives and organizational factors with types of consumer collaboration for innovation. This answers the call for more extensive research on internal success factors and barriers for open innovation that has been put forward in several recently published studies (e.g. Bogers et al. 2017; Foss et al. 2011; Gemser & Perks 2015; Lüttgens et al. 2014; Mahr et al. 2014; Randhawa et al. 2016; West & Bogers 2017). Depending on the channel co-creating organizations use to connect with consumers, the locus of innovation is located closer to or further outside the boundaries of the focal firm. We therefore expect there to be different degrees of formalization of the focal firm s co-creation processes, spanning from using Social Media channels to working with open innovation intermediaries and finally to establishing firmhosted innovation platforms as the most institutionalized form of consumer cocreation. We hypothesize that companies are collaborating much more intensively with consumers if they build a designated open innovation platform. Methodology To get a clearer picture of the landscape for open innovation with consumers in the consumer goods sector, we are collecting data via an online open-response survey directed to innovation and R&D managers as well as marketing managers. At first, our research focuses on the German market. Both large multinational companies and SMEs will be included. While there are many single or multiple case studies on the topic of open innovation with customers and users, so far, to our best knowledge, there has been no systematic examination of how companies formalize their consumer co-creation initiatives based on survey data derived directly from enterprises. We are conducting a survey with open-ended questions, as this is recognized as a well-suited methodology for identifying the most salient issues in a specific area of interest (Geer 1991). The survey is based on an extensive literature review that has established the relevance of the topic. It consists of five parts asking for 1. General information about the company and the respondents position 2. the existence of open innovation initiatives and the type of approach taken to 61

75 innovate with consumers 3. more specific information on the characteristics of the initiatives 4. the reasons for engaging in different types of open innovation with consumers, the expected results and perceived project success 5. the internal processes developed to support the co-creation initiatives, including experienced difficulties Expected results and implications In addition to identifying the most widely used channels for and the degree of institutionalization of customer co-creation in companies of different sizes across different consumer goods industries, we hope to gain in-depth insights into internal decision processes as well as organizational changes that need to be made when firms increasingly engage in co-creation. Our findings lay the groundwork for developing a research model for a large-scale quantitative survey analysing organizational antecedents of consumer co-creation. This second study will be conducted at a later stage of the research project to empirically validate the results from the open-response survey. By July 1st, 2017 we will have finished the data collection for the open-response survey and expect to present first results on the extent of institutionalization of customer co-creation for innovation in the German consumer goods industry. Our results complement conceptual and case-study based research on typologies of virtual co-creation, platform models, and innovation intermediaries (e.g. Piller et al. 2011; Keinz et al. 2012; Lüttgens et al. 2014; Rohrbeck et al. 2010) by integrating analyses of the implementation of co-innovation initiatives via different digital channels based on data collected directly from companies. 62

76 Investigating Self-Selection Mechanisms Induced by Selective Revealing Poster Markus Deimel (Vienna University of Economics & Business), Christopher Lettl (Vienna University of Economics & Business) Collaborations between organizations that are distant with regard to their knowledge bases can be valuable. Knowledge recombination enables partners to create novel products and services (cf. Hargadon, 2003; Nooteboom, 1999). Nevertheless, organizations often search for and form ties with partners that possess similar knowledge (Stuart & Podolny, 1996) because information regarding the collaboration partner is available, for instance, trustworthiness (Gulati, 1998) or resource endowment (Mitsuhashi & Greve, 2009). In addition, search and evaluation of a smaller number of potential partners is less expensive (cf. Afuah & Tucci, 2012; Katila & Ahuja, 2002). Given the potential value of collaborations with partners of different knowledge, how can organizations overcome local search and find distant collaboration partners? Alexy, George and Salter (2013) suggested selective revealing of knowledge as an approach to find distant collaboration partners. Selective revealing is the voluntary, purposeful, and irrevocable disclosure of specifically selected resources, usually knowledge based, which the firm could have otherwise kept proprietary, so that they become available to a large share or even all of the general public, including competitors. (Alexy et al., 2013, p. 272) With this approach, the focal organization makes some part of its knowledge available. This knowledge can be posted as solutions on the company website or a web-based platform. Being accessible, other organizations could then find these solutions. This is not unlikely as so called boundary-spanners, gatekeepers or technology scouts of other organizations regularly scan the environment for useful knowledge (Allen, 1977; Birkinshaw, Bessant, & Delbridge, 2007; Hargadon, 2003; Tushman, 1977). Selective revealing as a candidate approach to identify distant partners relies on the ability of the revealing organization to formulate the solution in a way that facilitates interpretation and assessment by distant actors (von Krogh et al.,

77 quoted in Alexy et al., 2013; Baer, Dirks, & Nickerson, 2013). In turn, actors need to recognize shared knowledge that is useful for them. Solutions from distant domains may contain knowledge for the actor that is more difficult to interpret due to its novelty (Boudreau, Guinan, Lakhani, & Riedl, 2016). These difficulties manifest themselves in false negatives or false positives. In the former case, organizations may overlook solutions even though the collaboration could be valuable. In the latter case, a solution from a different domain may be appealing at first, but results in lost time and resources as inefficiencies or incongruences become apparent. Therefore, appropriate formulation of the solution as well as accurate recognition of useful solutions by technology scouts become crucial determinants both for the revealing and receiving organizations. Alexy, George and Salter (2013) distinguish between the disclosure of solutions and problems as mechanisms to find collaboration partners, the latter also known as crowdsourcing. Recent research in the context of crowdsourcing problems investigated the role of formulation in terms of distance of the content of the solution to an organization s problem (Piezunka & Dahlander, 2015) and in terms of problem characteristics so that potential providers of solution knowledge allocate attention to the problem (Haas, Criscuolo, & George, 2015; Piezunka & Dahlander, 2015). Researchers also appended the technical descriptions of problems with need-related information (Ehls, Fretschner, Lakhani, & Lüthje, 2016). Foss, Frederiksen, & Rullani (2015) showed that programmers change their behavior, when exposed to structured or less structured communication in open source projects. In essence, despite some projects that explicitly target formulation, factors such as the complexity of the knowledge, abstraction level or status remain unexplored. The strategy literature portrays partner selection as a long and difficult search process that includes strategic and political criteria (Dodgson, 1993; Geringer, 1988; Lorange & Roos, 1992). Evaluation of potential partners focuses on the simultaneous assessment of various partner attributes including knowledge (Tyler & Steensma, 1995). We do not know much about the evaluation of a potential partners knowledge in isolation. Anonymous postings of solution knowledge on webbased platforms, however, are common in the context of selective revealing. Strategy researchers have conceptualized approaches to build networks with unknown partners based on selective revealing (Fjeldstad, Snow, Miles, & Lettl, 2012). Others proposed that technology scouts play an important role when 64

78 organizations search for new connections with distant partners (Birkinshaw et al., 2007). Yet, apart from qualitative studies (e.g. Garud, Jain, & Kumaraswamy, 2002), there is little empirical evidence about the benefits of selective revealing as an approach to form collaborations. Research questions Thus, we formulate the research questions: How does the formulation of solutions influence the selfselection of individuals with distant knowledge? Which characteristics of distant individuals enable them to recognize the usefulness of revealed knowledge? This project follows the call of Alexy et al. (2013) for research regarding the role of formulation in selective revealing. It will shed light on the benefits of revealing knowledge in a very early phase of search for collaboration partners. Given that distant technology scouts find the knowledge and contact the focal firm, the focal firms search can include potential partners with distant knowledge. For managers, we will depict how formulations of technological knowledge affect whether distant actors recognize the shared solution as an opportunity or not. With regard to these actors that look to profit from shared solutions, we will determine factors that influence their ability to recognize potentially useful information. We plan to conduct a within-subject repeated measures experiment. This design allows to focus on the effects of solution formulations and to reduce noise due to differences between individuals. To increase the external validity of the experiment, we will sample participants from different knowledge fields. This design should also help us to investigate the interaction between knowledge distance and solution formulations. We complement the experiment with a survey to extract characteristics of distant individuals that make them better in recognizing the usefulness of revealed knowledge. To date we are preparing pre-tests for our study and would like to share some first insights at the Open and User Innovation Conference in Innsbruck. 65

79 Ecosystems in Idea Evaluation. An Explorative Analysis of Multiple Crowdsourcing Platforms Poster Ralph Lichtner (Freie Universität Berlin) Pure user or collective idea evaluations with subsequent process phases represent a heterogeneous phenomenon on crowdsourcing platforms due to different company and intermediary approaches. The conception of the evaluation phase is paramount for the best idea choice, ultimately resulting in a better chance for innovative products and services. However, until now open and user innovation research does not progress with a broad comparative and exploratory study on the ecosystems of joint idea evaluation. Therefore, this present research approach includes a holistic comparative analysis of highly active intermediary and company-owned crowdsourcing platforms, most of them are strongly user-driven, to ameliorate open and user innovations research streams in an up-to-date precise manner by focusing on the early joint idea evaluation phase. By examining the constituent components for the processes the evaluation mode, the composition of the evaluation teams, the team size, the duration of the evaluation process, the evaluation criteria, the process of user involvement, and the composition of the decision makers characteristic patterns of an efficient and effective evaluation method for high-quality ideas can be identified and compared. This is achieved through an explorative online analysis of 22 highly active crowdsourcing platforms. The first carrying insights show that the design of multi-stage evaluation processes with appropriate timebased integration reflect a conceptional trend for crowdsourcing platforms with high and active user involvement. Additionally, depending on the crowdsourcing type, first insights show: (1) how and when users are included or excluded in the actual idea evaluation and when users are only involved in the "pre-filtering", (2) which resource-conserving aspects are important from a company perspective and (3) how active community involvement of users might be coupled into the evaluation processes with additional motivational factors. Based on data of the selected 22 platforms, this research contributes to the 66

80 understanding of patterns regarding team composition and size of user-, company-, and external experts in the idea evaluation process. The next research steps include a focused and balanced analysis on the evaluation process through company-wide expert interviews ameliorated by additional user expert interviews. 67

81 Idea-Related Knowledge Accumulation in Online Innovation Collectives Poster Brita Schemmann (Utrecht University), Maryse M.H. Chappin (Utrecht University), Andrea M. Herrmann (Utrecht University), Gaston J. Heimeriks (Utrecht University) The Internet has enabled new forms of knowledge production and open innovation processes in online innovation collectives. Understanding what facilitates idea elaboration and development in online idea calls is crucial to make better use of the crowdsourcing potential. Our findings therefore seek to contribute to a better understanding of knowledge accumulation in an unstructured, uncertain and fluid knowledge production environment, i.e. in those online innovation collectives that invite a diverse crowd with no specific expertise to individually come up with a wide range of ideas or solutions for fairly broad topic. We therefore assess what influences idea-related knowledge accumulation in the commenting threads. Using data collected from an idea crowdsourcing call we apply a cross-sectional research design to assess the influence of (1) factors related to the relative advantage of the initial idea and (2) factors related to the interaction in the commenting thread. Our first results show that idea-related knowledge accumulation does take place in these fluid online innovation collectives, but that it cannot be taken for granted. They indicate that different factors influence the amount of knowledge accumulation taking place in a thread. Factors related to the relative advantage of the initial idea (both novelty and perceived usefulness) indicate to have significant positive effects. Moreover, the interaction among commenters seems to significantly influence idea-related knowledge accumulation: while a negative seed comment has a negative influence on the amount of knowledge being accumulated in the thread, direct interaction between commenters has a positive effect. Concerning the latter it does not seem to matter whether the comments that directly refer to other commenters' input mainly value or question the other input. We 68

82 observe no significant effect for the activation of input via questions and invoking comments. 69

83 Evaluating Crowd-Sourced Ideas: Influence of Idea Characteristics on Crowd and Organizational Evaluators Poster Lisa Kristina Wimbauer, (University of Passau), Patrick Figge (University of Passau), Carolin Häussler (University of Passau) Organizations engage with crowds to generate novel ideas (Afuah & Tucci, 2012). However, most research does not consider which ideas are actually realized. Idea selection is challenging since a large number of highly diverse submissions have to be reviewed. Organizational evaluators may not possess expertise in all knowledge areas and lack the time to adequately evaluate all ideas (Hienerth & Rinar, 2013). For instance, an increasing number of ideas narrows the attentional focus of evaluators, who then pay less attention to novel submissions (Piezunka & Dahlander, 2015). This thwarts the benefits that organizations set out to capture. Thus, crowds diversity may be used to support the judgement of organizational evaluators if they apply the same selection criteria. Conversely, if the crowd s decisions are based on other criteria, they may be used to correct certain selection tendencies (King & Lakhani, 2013). There are only few studies comparing evaluations by crowd and organizational evaluators and even fewer consider reasons for differences between them. For instance, crowd members pay more attention to less novel ideas (Haas, Criscuolo, & George, 2015), whereas organizational experts favour projects with medium levels of novelty (Criscuolo, Dahlander, Grohsjean, & Salter, 2016). However, most studies do not address differences between organizational and crowd evaluators. Moreover, prior work focuses on a limited number of idea characteristics and is usually not in the context of crowd-sourced ideas. Understanding how ideas with different characteristics are assessed by crowd and organizational evaluators paves the way to make better use of crowds in idea selection. We conduct a large-scale study on how crowd-sourced ideas differentiated by their content, linguistics and presentation are evaluated. Our research question is as follows: RQ1: How do idea characteristics influence idea evaluation? Are there differences between the crowd and organizational 70

84 evaluators? Rather than treating evaluators as uniform, we consider withingroup differences to further improve our understanding of crowd and organizational evaluators. We argue that individual characteristics influence their evaluations. Prior research predominantly focuses on the influence of individuals expertise. For instance, Jeppesen and Lakhani (2010) find that crowd members with expertise from areas distant to the focal field of the problem provide the most valuable solution. Magnusson, Waestlund and Netz (2016) report that technically naïve crowd members rate originality, user value and feasibility of ideas higher than technical skilled crowd members and experts. There is only limited research on how idea characteristics are perceived differently by individual crowd and organizational evaluators. With the intention to benefit from the high number of crowd evaluators and the diversity of their knowledge, we make an effort to go beyond merely considering their average opinion. Thus, we investigate: RQ2: How do individual characteristics of crowd members and individual characteristics of organizational evaluators influence idea evaluation? We investigate data from crowdsourcing projects of the BMW Group Co-Creation Lab, a platform in cooperation with the innovation company HYVE. Our dataset contains over crowdsourced ideas generated by about 500 crowd members in three innovation contests during the years We have detailed information on idea and individual characteristics of 15 organizational evaluators and over 800 active crowd members. The crowd evaluated ideas using a like and a use button and free text comments. Organizational evaluators rated ideas on four criteria (novelty, customer value, company fit and overall impression). To capture differences between the ideas, we operationalize several variables for each of these three characteristics: the ideas content, linguistics and representation. The content is coded manually by assigning categories to all texts written by idea generators. We used Linguistic Inquiry and Word Count (LIWC), a validated dictionary-based method of computerized text analysis, to extract the meaning from how texts are written. We generate measures that reflect emotional and analytical tone or the cognitive orientation. For idea representation, we distinguish text length, visualizations and completeness. Individual characteristics describe the socioeconomic background (e.g. education, professional experience), product affinity (e.g. number and brand of cars owned), platform activity (e.g. idea contributions, contest participation) and for organizational evaluators their 71

85 position in the organization (e.g. hierarchy, department affiliation). Preliminary First analyses suggest that idea representation (e.g. idea completeness, idea length) influences both, crowd and organizational evaluators. However, they seem to differ especially in how they are influenced by linguistic characteristics (e.g. emotional vs. analytic tone). The manual coding of the ideas content is currently being finalized. By July 2017 we will present empirical analyses on the differences between crowd and organizational evaluators for each of our three idea characteristics (content, linguistics, representation). Further, we will present analyses on how characteristics of crowd and organizational evaluators lead to differing evaluations within each group of evaluators. We expect interesting contrasts between the two groups of evaluators and within each group of evaluators, not only for evaluations at the mean, but particularly at the top and bottom percentiles of ideas. For instance, organizational and crowd evaluators may be aligned when it comes to sorting out low quality ideas, while they identify different ideas to be among the top 5% or 10%. 72

86 Solver Motivations in Complex Innovation Contests Poster Ademir Vrolijk (The George Washington University), Zoe Szajnfarber (The George Washington University) An important part of designing innovation contests, where solvers compete to achieve a predetermined technology goal, is setting the correct incentives for the contest. Solvers are, in turn, drawn to these contests by the potential outcomes that they see for themselves after deciding to participate. Understanding the link between contest outcomes and the related (external) motivations can form a powerful lever for the seeker to drive the needed participation. Despite the long history and wide ranging topics covered by these contests (from the relatively simple to very complex), this link is not well captured in innovation contest theory. Empirical studies show active solvers more motivated by pecuniary incentives, but other times more motivated by non-pecuniary ones or making investments in their submissions that far exceed the total prize award. By not digging into the dynamics of what this link entails, future seekers could misapplying incentives, eliciting participation from a nonoptimal (part of) the crowd. As such, scholars have recognized this gap and have called for further study of motivations of solvers and their links to the incentives across the range of contests. We address this gap through an inductive study of solvers in National Aeronautics and Space Administration s complex innovation contest platform, Centennial Challenges. The innovation contests launched by this platform range in complexity (prizes range from $50k to $5M), and participation (solvers include academics, hobbyists, and industry). We performed semi-structured interviews with approximately 40 active solvers, and their motivations were open coded from the transcripts of these conversations. Iterating between the literature and the data, our codes were combined into constructs that revealed patterns of motivations. Our preliminary observations show that at varying levels of complexity solvers view the same prize types (prize money, reputation gains, exposure) differently, which explains some of the conflicting observations reported previously. This stems from the way in which solvers view the contest in relation to other activities. In smaller-scale contests, 73

87 solvers often view the contest in isolation, contributing a one-off solution generated specifically in response to the contest. In contrast, in larger-scale contests, where the barriers to any contribution tend to be higher, solvers tend to contribute when they can leverage ongoing work to generate their solution. Viewed this way, they benefit from making progress on ongoing projects whether or not they "win" the contest. We are further developing our findings and expect to have a completed analysis by July This insight has implications both for how participation should be incentivized and who tends to participate. With respect to incentives, we note a shift in importance away from innovator-centric to technology or project-centric as complexity increases. With respect to participation, our observations suggest that an emphasis on drawing diverse contributors may be less productive in complex, large scale prizes. Combined, these imply that complex problems, which require expert capabilities and knowledge to solve, should draw potential participants through technology or project-centric incentives to maximize their potential for success. 74

88 The Role of the Entrepreneur in Managing Crowdsourcing in SMEs Poster Izabella Bereczki (Universität Innsbruck), Despite the increasing number of literature about crowdsourcing, little is known about the obstacles, entrepreneurs face, while they manage crowdsourcing. Small and medium-sized enterprises (SMEs) benefit from crowdsourcing, but most of them still have difficulties in managing the process. In order to understand the SMEs innovation strategy, we have to link entrepreneurship to open innovation. The combination of the existing literature about entrepreneurship and crowdsourcing will serve as a theoretical foundation for my research. Keywords: crowdsourcing, SME, entrepreneurship Introduction An empirical study should provide information about the attempts innovative SMEs made in order, to solve the problems they encountered. Furthermore, it is important to find out what the entrepreneur can do to increase the value they get from the crowd. The entrepreneur has a key role in the SMEs, and probably in the crowdsourcing process as well. His or her vision, innovative, management and problem solving skills lead the enterprise towards success or failure. In the meantime, the reactions of the employees affected by crowdsourcing can be diverse. Some of them might feel threatened, and this could cause tensions. However, others could view crowdsourcing as an opportunity. The research study will focus on the directly affected employees, and hopefully reveal the entrepreneurial challenges in managing the process. Methodology A qualitative research approach is planned, to achieve the research goals. Qualitative methods are valuable here because, it might capture the participant s perception of crowdsourcing, and the possible internal conflicts. Case study research strategy will be applied. Furthermore, the data collection will consist of semi structured interviews, observations and other available and relevant sources. 75

89 Who are Your Design Heroes? Exploring User Roles and Behavior in a Community-Based Design Contest Poster Manuel Moritz (Helmut Schmidt University), Tobias Redlich (Helmut Schmidt University, Jens Wulfsberg (Helmut Schmidt University), Community-based online innovation contests are very popular and effective means for idea generation and problem solving for both sides, organizations and users. Participants from all over the world with very diverse backgrounds and areas of expertise are invited to submit their ideas and solutions to a specific task or problem. Communities and contests have been studied by scholars of many research fields such as innovation management, management sciences as well as economics. Subsequently, there is a rich body of literature from different disciplines on the design of contests (awards, duration, popularity etc.), on users (attraction, motivation, communication, benefits) as well as on innovative outcomes. Interestingly, recent research found that people not only compete in challenges to win a prize, but also socially interact and collaborate with other users of the contest community, e.g. by commenting, giving feedback and exchanging ideas. Community-based contests that rely on competitive and collaborative elements at the same time ( communitition Hutter et al., 2011) thus represent a promising contest mode with respect to attraction, motivation and outcome. Little research, however, focused on this specific configuration. Building on Füller et al. (2014) who found specific roles regarding network position and communication behavior in a community-based contest, this research approach attempts to add to the body of literature by examining users and their specific roles and behavior of a different contest setting. Better understanding participants and their behavior helps to properly design and manage innovation contests and to build up and sustain a community of innovative users for successful co-creation. The US-based tech company Local Motors (LM) managed to build up a co-creation community of about enthusiasts who collaborate along open source principles and by means of webbased toolkits on its platform. Besides co-creating with users for their own 76

90 product portfolio, LM also hosts and manages challenges in partnership or sponsorship with other organizations that want to make use of co-creation tapping the LM community (e.g. Urban Mobility Challenge, Domino s Pizza Ultimate Delivery Vehicle Challenge). In this case, LM acts as an intermediary between the seeker who specifies the task and sets the challenge conditions (IP regime, prize, evaluation) and the solvers from their community, like the concept of crowdsourcing. In 2016, LM launched another design contest on its co-creation platform: the Airbus Cargo Drone Challenge (ACDC) in partnership with Airbus Group. People were invited to submit design concepts for a commercial drone to build upon Airbus Quadcruiser concept. The concept had to meet certain specifications regarding design, size, weight, payload, operation mode etc. to perform the task of quickly delivering urgent medical supplies in case of an emergency or a disaster. Each entry has its own project page where all information about the concept (text, design, drawings etc.) is posted and people may add comments. In total, contributors could win a $ 117,500 cash prize that was awarded in three distinct categories. In the end, 425 entries were uploaded and publicly revealed on the LM platform. The case of LM and the ACDC in particular is of special interest for various reasons: First, the community which operates in a technology niche (industrial design) entails like-minded enthusiasts and subject matter experts (engineers, designers, entrepreneurs) rather than consumers or customers. It is comparable to an open source community where highly skilled people jointly and collaboratively tackle problems and develop products. In fact, most of the projects on the LM platform were initiated by users. Even if LM hosts competitive ( the-winner-takes-all ) contests on the platform, it still represents a collaborative co-creation environment. All posts and project or challenge related entries are publicly available and a Creative Commons license is used to encourage knowledge sharing and idea exchange. People may comment on entries, upload improvements, and discuss in forums. So, the LM community seems to be co-opetitive by nature. Additionally, the community comprises existing as well as new members during a challenge. First, an exploratory survey (N=81) was conducted to find out more about the heterogeneity of the users, their motivation and behavior. We found that the users live all over the world, are highly educated, represent a wide range of income and employment situations and majorly perceive themselves as engineers and designers. Users are predominantly motivated by intrinsic factors 77

91 such as learning, having fun and solving problems. They release their work early and are heavily engaged in commenting and feedback. The next steps (with results expected by July 1st) are: social network and cluster analysis (536 nodes, 1800 edges) as well as interpretative content analysis of comments to identify distinct user roles based on contribution behavior (comments, quality and number of designs). This research design was found to be very effective for studying user roles. 78

92 Crowdfunding 79

93 Beyond the Platform - Designing Enterprise Crowdfunding to Foster Intrapreneurship Paper Robert Kleinscheck (Technical University of Munich), Katja Hutter (University of Salzburg), Johann Füller (Universität Innsbruck) Fostering the innovative potential of entrepreneurial employees also referred to as intrapreneurs becomes a crucial competitive advantage for established companies. While intrapreneurs have shown to have positive effects on R&D efforts and finally on firms performances, their occurrence is not self-evident. Adequate organizational antecedents, processes and tools are necessary to support them. This study follows a design science research approach and conceptualized, created and implemented an enterprise crowdfunding tool in a company to foster intrapreneurship. Interviews with users and managers were conducted to explore which design principles for the corporate application of a crowdfunding platform support employees to act entrepreneurial in the fuzzy front-end of innovation. The resulting implementation features and design principles provide new insights to crowdsourcing and intrapreneurship literature and reveal suggestions how enterprise crowdfunding can be implemented to foster intrapreneurship. 80

94 Does Hierarchical Distance Bias Idea Evaluation? Paper Tim G. Schweisfurth (Technical University of Munich), Michael A. Zaggl (Technical University of Munich), Claus P. Schöttl (Technical University of Munich), Christina Raasch (Technical University of Munich) The selection of ideas for developing new products or services is an increasingly relevant challenge for many organizations. Research shows that idea selection is often affected by biases, which lead to the problem that not the best ideas are selected. For example, it has been shown that ideas are systematically undervalued if the idea evaluator comes from another department than the idea creator. Building on the notion of homophily, we theorize that hierarchical similarity can produce another evaluation bias. Specifically, we expect overvaluation of ideas when evaluator and creator are hierarchically similar, i.e., they are on the same hierarchical level or close to each other. As alternative concepts to homophily, we consider social status and competition. We test our prediction with a unique dataset from an idea crowdfunding contest in an internationally operating company (Siemens). In the contest, employees could create ideas and then evaluate them by investing into them. As theorized, our results show that ideas are overvalued (undervalued) when evaluator and creator are hierarchically similar (distant). We further show that idea novelty amplifies this effect. Our results also support prior empirical research showing that ideas from the same business unit are overvalued, but we did not find evidence for gender homophily. We conclude our paper by outlining possibilities for future research and providing implications for theory and for organizations who consider enterprise crowdfunding for selecting ideas. 81

95 Crowdfunding Science? Poster Lars Frederiksen (Aarhus University), How is knowledge production through scientific research best funded? New digital organizational forms serve as experiments for how crowds can fund science in faster and perhaps more democratic ways. Still, we know little about which types of research crowds invest in and why? Is it different from research sponsored from competitions initiated by traditional research councils? Why do crowds and experts not value the same type of research and thus exercise different selection heuristics? Who asks for crowdfunding for research and how does obtaining funding for research from the crowd influence the career trajectory of the applying researcher? Is crowdfunded research published and thus does crowdfunding science produce scientific knowledge? Who are the backers of crowdfunded research and does it matter for the validity of the crowdfunding research projects? The questions are many! Elements such as: project discipline, topics, pictures and description as well as status, gender, affiliation and social networks of project entrepreneurs may all play a part as possible explanations. This explorative research program aims over a series of three studies to extend our knowledge of the opportunities and possible threats from crowdfunding scientific research. 82

96 Crowdfunding for Family Firms in a Resource Exchanging Context Poster Martin Danler (Management Center Innsbruck) This poster will present the research proposal for my PhD Management study at the University of Innsbruck. It deals with the topic of crowdfunding in a resource exchanging context and investigates the willingness of different actors to back a campaign not only with financial resources. Family enterprises exist in a wide variety of sizes and forms of organizations, from small and medium-sized enterprises to multinational corporations (Zehrer and Siller 2007, Klein 2000, Getz and Carlsen 2005). In Austria, 425,276 enterprises are classified as SMEs and are said to be the backbone of the domestic economy (Hennerkes et al., 2007, Czernich et al 2005, EUBusiness 2009, 2009, Wirtschaftskammer Österreich 2014). In Tyrol, 99.7% of the companies are classified as SMEs, of which one-person enterprises are 51.3% and micro-enterprises 38.1%, with 46,400 employees representing the bulk of the economic structure (Office of the Tyrolean Regional Government 2013). These figures illustrate the considerable contribution of micro-enterprises to the gross value creation of a country (Astrachan and Shanker 2003, Culkin and Smith 2000, Klein 2004) and their importance for the continuity of an economy (Benckendorff and Zehrer 2013; Klein 2000). Several studies have shown that family businesses prefer low levels of debt and are therefore less risky (Anderson et al 2012, Le Breton-Miller et al., 2011, Muñoz- Bullón and Sanchez-Bueno, 2011; Lee 2013). A lower level of investment in research and development has also been found among family businesses (González et al., 2013, Mishra and McConaughy 1999, Welsh and Zellweger 2010, Zellweger and Sieger 2012), which also suggests that these companies are less innovative. Due to the risk aversion of family businesses, this study investigates crowdfunding as an alternative financing option. A first empirical survey should show whether family firms do have the readiness for crowdfunding under consideration of all four types of crowdfunding (Moritz and Block 2014): Donation based, Reward based, Lending based and Equity based. In a second step, the resource exchange model, developed by Foa and Foa (2012), is applied in order to investigate an actor s willingness to share not- 83

97 financial resources in a crowdfunding campaign. The model (see Figure 1 Resource Exchange Theory Model Foa & Foa (2012)) comprises six resources which can be exchanged between actors. These resources are Love, Service, Goods, Money, Information, and Status. The arrangement of these categories has a symbolic meaning, for example, the distance between Money and Love is broader than Money and Goods and thus more difficult to exchange than Money and Goods. Therefore, resources that are closer together are a lot easier to exchange than more remote resources. Further, the diagram shows not only the interrelationships between the resources but also those resources that are more likely to be exchanged. Human actions can involve several resources, but always with the principle of proximity. Within this model it is possible to analyze the exchange process in crowdfunding campaigns (Greenberg and Gerber 2014). The research questions for that proposal are Which kind of resources are different actors of a crowdfunding campaign willing to exchange? What kind of crowdfunding fits best for family firms and is most likely to be accepted? Data collection is carried out using an interview guide. This type of interview is characterized as a type of non-standardized interviews in which the interviewer uses a prepared list of open questions (the guide) as a basis for conversation. In particular, the interviews are recommended when a number of different topics are addressed in an interview which is not determined by the answers given by the interviewer, but by the purpose of the study. In addition, this form of the interview is useful when individual, precisely determinable information must be collected (Gläser and Laudel 2010). This form of empiricism is thus suitable for achieving the set of goals. The topic of crowdfunding can thus be discussed in detail and more detailed questions are possible through the partly structured questions. The different ways of thinking of the family entrepreneurs can also be discussed in this interview. After conducting the interviews, the interviews are transcribed. After completion of the transcription, the content of the collected material will be analyzed (Mayring 2008). In addition to the interviews, a student-based survey will be conducted by asking them for possible resource exchanges within a fictive crowdfunding campaign. For this fictive campaign students can choose different ways to contribute to this campaign, for instance to back a project with money, man power (service), suggestions (information) or goods and value those kind of exchange on a Likert-scale. The results of this survey should be evaluated with SPSS. Due to the lack of data on family 84

98 businesses in Tyrol we expect valuable data on crowdfunding in family businesses. In addition to answer the research questions, the result of this applied research project is a potential analysis for crowdfunding in Tyrol. In addition, the aim is to look at how a possible process for successful crowdfunding in family businesses could look like. If the result of this study shows that there is the readiness to exchange also non-financial resources, then we found a new way to back crowdfunding projects what additionally could increase the probability for more successful crowdfunding campaigns. 85

99 Beyond Funding Exploring the Effect of Enterprise Crowdfunding on Employee Engagement Poster Carina Benz (Karlsruhe Institute of Technology), Niels Feldmann (Karlsruhe Institute of Technology) Crowdfunding has become a global phenomenon, using the public to obtain financial support for projects or ventures. Recently, it has drawn increasing attention among companies and public organizations for its possible application in innovation management. Organizations are experimenting with the in-house usage of the crowdfunding mechanism to drive innovation: In so-called enterprise crowdfunding (ECF), they enable employees to propose project ideas and to fiduciary invest company funds in idea proposals submitted by their coworkers. While the core of research related to (enterprise) crowdfunding focuses on the idea generation and the evaluation effect of the mechanism, we are interested in the presence of spillover effects on employee behavior and organizational culture. The existence of spillover effects of ECF and related Enterprise 2.0 mechanisms on the social economic environment is suggested by literature as well as they are quoted as an explicit objective by executives (Agrawal, Catalini, & Goldfarb, 2014) (Muller, Geyer, Soule, Daniels, & Cheng, 2013) (Richter, Stocker, Müller, & Avram, 2013). In this context, we concentrate our study on the ECF effect on employee engagement as one possible spillover effect among many (like impacts on cross-silo collaboration, intrapreneurship culture and others). The concept of employee engagement was initially defined as the investment of a person s preferred self in work-related tasks, in physical, cognitive and emotional presence, as well as in active role performance (Kahn, 1990). Being connoted with attributes such as passion, enthusiasm, activation and energy, high levels of employee engagement are generally considered as a desirable state for organizations (Macey & Schneider, 2008). Although the concept is primarily discussed in organizational contexts as general work engagement, it may as well exert a positive impact on innovation activities and creativity (Schohat & Vigoda-Gadot, 2010). Employee engagement is thus 86

100 interpreted in the sense of employees feeling involved and proactively participating in the company s innovation activities (Brodie, Hollebeek, Juric, & Ilic, 2011). We aim to study whether the application of enterprise crowdfunding in an organization influences its employees engagement. To explore this particular ECF spillover effect, we are currently conducting an empirical study at a global manufacturing and electronics corporation surveying both participants and non-participants of an ECF run. Following Macey and Schneider (2008), we distinguish three sub-concepts of employee engagement: (1) trait engagement, (2) state engagement, and (3) behavioral engagement. As the applicability of existing surveys (e.g. Utrecht Work Engagement Scale, ISA Engagement Scale) for studying engagement in the context of innovation management is limited due to their focus on general work engagement, a dedicated survey instrument based on the Macey and Schneider (2008) framework is designed. Each of their three concepts is represented by four questions: E.g., state engagement is covered by questions considering satisfaction, involvement, commitment, and empowerment in innovation. Data is collected at three points in time (before the ECF run, directly after the ECF run and 3 months after the ECF run). 400 employees pre-registered for the current ECF run as well as a control group of 400 unregistered employees are asked to answer the survey at all three points in time. By complementing quantitative survey data with qualitative interviews with ECF participant, we aim to provide additional validity to the study. Early statistics based on 208 responses suggest a statistically significant difference between participants and non-participants state engagement measured at the point before the ECF run. At OUI, we will present findings from data collected at two points in time allowing a comparison of employee engagement before and after enterprise crowdfunding as well as between the group of ECF participants and the control group. With this study, we aim to contribute to an expanded research horizon for market and innovation management mechanisms, such as ECF, by investigating them from an organizational behaviour and innovation climate perspective. In terms of practical implications, potential positive results of our study may broaden the view on benefits that can be gained from the internal application of collective intelligence approaches like ECF instruments. They are potentially powerful tools for shaping organizations innovation capabilities - way beyond mere good decisions. 87

101 Do Consumers Really Care About Crowdfunding? Exploring the Crowdfunding Effect on Innovation Perception Poster Christian V. Baccarella (Friedrich-Alexander-Universität Erlangen-Nürnberg), Timm F. Wagner (Friedrich-Alexander-Universität Erlangen-Nürnberg), Kai-Ingo Voigt (Friedrich-Alexander-Universität Erlangen-Nürnberg) In recent years, crowdfunding has gained significant attention as an innovative way of online-based fundraising (Belleflamme et al. 2014). In contrast to traditional funding instruments, crowdfunding is a collective effort with a large number of individuals involved and with high visibility of the funding progress on various crowdfunding platforms (Colombo et al., 2015). Several crowdfunded products have been widely discussed in the media (e.g. the pebble watch) and have attracted considerable amount of press attention (Mollick, 2014). The increased public interest and hype about crowdfunding leads to the assumption that the mere fact that a product has been successfully crowdfunded may serve as a differentiator that could positively influence consumer attitudes. Therefore, we seek to explore whether the crowdfunding label as such has the potential to create more favorable consumer attitudes. In the following, we present two out of three experiments that extent the understanding of consumer perceptions in the context of crowdfunded products. In Study 1, we sought to explore whether the fact that a product was financed through a crowdfunding campaign has an influence on consumers innovation and brand perceptions. An experimental study was designed, using a 2 (crowdfunding: yes vs. no) 3 (product category: smart light bulb vs. smart earphones vs. electronic nail clipper) between-subject design, resulting in six experimental conditions. For each product category, two advertisements were designed: one including a small note that the product was financed through a crowdfunding campaign, and one without this note. Besides the difference regarding this crowdfunding note, the ads were identical for each product category. On the left side of the ads, a headline and a short copy text introduced the participants to the product. On the right side, the ads presented a picture of the product, demonstrating its 88

102 functionality. An online questionnaire was distributed and participants were randomly assigned to one of the six conditions. Before participants were exposed to one the six ads, they were briefly introduced to the fictitious brand Tanabo. In the non-crowdfunding condition, Tanabo followed a common funding approach ( external investor ). In the crowdfunding condition, Tanabo used Kickstarter to finance its product idea. Here, we included a very short and easy description of crowdfunding to ensure that everyone understood the basic idea of crowdfunding. After the brand description and the product ad, participants reported the dependent variables (attitude toward the brand, attitude toward the product, innovation ability of the brand, consumer confidence in the brand, recommendation intent, purchase intent; see Appendix) and the control variables (product involvement, consumer innovativeness, crowdfunding familiarity, crowdfunding mindset, age; see Appendix). Data collection resulted in 282 valid questionnaires (55.0% female, average 28.5 years). To isolate the main effect of crowdfunding on consumer responses, we collapsed the sample across the three product categories. A multivariate analysis of covariance (MANCOVA), including the control variables as covariates, revealed that there are no significant effects of crowdfunding on the dependent variables (ps >.1), besides a weak significant effect on consumer confidence (p <.1), indicating that participants in the crowdfunding condition have a higher confidence in the brand that it can deliver a satisfactory product than participants in the non-crowdfunding condition. Study 2 had two objectives. First, we attempted to replicate the somehow surprising results of Study 1 by using other product categories. Second, we sought to explore the moderating role of product innovativeness on the effect of crowdfunding on consumer perceptions. We followed a similar experimental approach like in Study 1. Yet, two differences were included. First, we used different product categories. Second, for each product category, we created a highly innovative version and an incrementally innovative version. In other words, we conducted a 2 (crowdfunding: yes vs. no) 3 (product category: suitcase vs. sport shirt vs. pen) 2 (product innovativeness: low vs. high) between-subject design, resulting in twelve experimental conditions. Product innovativeness was manipulated by presenting different technologies to the study participants. For example, the suitcase ad in the low-innovative condition presented a rolling technology (newly developed ball-bearing technology that ensures smooth and easy 89

103 rolling). On the other hand, the suitcase ad in the high-innovative condition presented an autonomous suitcase technology (specific location sensors and several electric motors enable the suitcase to autonomously follow its owner). A pretest was performed beforehand to ensure successful manipulation. For the pretest, 37 participants were introduced to four technologies (two lowinnovative and two high-innovative) for each of the three product categories. We used a 1-item 7-point differential scale (Not at all innovative / Extremely innovative) to assess the innovativeness of the technologies. Then, we used the technologies with the greatest differences in regard the innovativeness scale (ps <.01), resulting in the six technologies included in the main experiment. For the manipulation check, we used a more sophisticated scale for product innovativeness (see Appendix). Again, the dependent and control variables used in Study 1 were included. Data collection resulted in 576 valid questionnaires (47.4% female, average 37.0 years). Manipulation of product innovativeness was successful (p <.01). Again, the sample was collapsed across product categories. A MANCOVA, including the control variables as covariates, showed no significant effect of crowdfunding on our dependent variables. Additionally, the interaction of crowdfunding and product innovativeness was non-significant for all dependent variables (ps >.1). Although recent research has acknowledged the positive effect of crowdfunding for customers as funders (Bitterl & Schreier, 2017), the two experiments show that the crowdfunding label might generally not work as good as expected. Given the fact that only a small fraction of future customers actually participate in a crowdfunding campaign raises further concerns about the effectiveness of the crowdfunding label as a communication tool. However, our two experiments contained only a subtle manipulation between the crowdfunding and non-crowdfunding conditions. Therefore, our third experiment will provide a different perspective at the crowdfunding effect by integrating a stronger narrative element into the study design. We assume that the crowdfunding label may be especially effective when telling the whole story behind a product. 90

104 Firm s Interaction with User Innovation 91

105 Wicked Design for Wicked Goals - How Hyperloop Transportation Technologies Challenges Organisation Design Theory Paper Ann Majchrzak (University of Southern California), Terri Griffith (Santa Clara University), David Reetz (Technical University of Munich), Oliver Alexy (Technical University of Munich) Theories of problem-solving and organization design in profit-seeking organizations assume that organizations are structured to accomplish a specific goal. While there is literature among non- profit organizations about goal conflicts among stakeholders, and among R&D organizations about projectbased organizing, none of this considers how to design an organization when the goal of the organization requires growth by simultaneously tackling new markets, social movements, and novel technologies; the creation of a new sustainable ecosystems of diverse partners; and the use of the internet to acquire a worldwide network of expertise. Such an organization faces what we refer to as wicked goals, i.e., goals which are multifarious and proactive. We identify and study Hyperloop Transportation Technologies, Inc. (HTT), an organization with the wicked goal of changing transportation through crowdpowering. We show how HTT organizes around five design principles, having rejected traditional organization design: Branching Rather than Dividing, Opportunity Search for Boundary Porousness, High- touch Virtual Community, Disentangling Information Dependencies, and Bottom-up Cooperative Disruption. We label this Wicked Goal Design, which, at its core, creates a balance between generativity and structure to ensure openness and effectivity. Our insights into these design choices hold the potential to invigorate theorizing on around new organizational forms. 92

106 An Exploration of On-Demand Innovation Model in Organizations Paper Jiang Huang (Tsinghua University), Jin Chen (Tsinghua University) This study develops a microlevel and holistic framework showing how firms can integrate need and solution knowledge from both inside and outside the firm and then transfer this knowledge into products or services with the help of internal employees and external agents. Specifically, this study attempts to replace the traditional linear model of innovation with an interactive hourglass model of innovation. Accordingly, open and user innovation perspectives are adopted to explain the constitution and functionality of the new innovation model. The theoretical framework is illustrated in terms of key processes and several embedded cases from Haier, a firm that thrives in the competitive home appliance market by successfully using both full-time employees and boundary spanners, in addition to employing vast resources on a global scale. 93

107 The Digital Revolution, 3D Printing, and Innovation as Data Paper Aric Rindfleisch (University of Illinois), Matthew O'Hern (University of New Hampshire), Vishal Sachdev (University of Illinois) The digital revolution has created an extremely data-rich environment. Increasingly, firms are seeking to acquire and analyze a variety of consumer data such as online shopping, social media and web browsing behavior to enhance their innovation activities. We term this approach Innovation from Data. At the same time, a growing number of consumers are gaining the ability to transform digital data into innovative physical products through the use of new tools such as audio and video editing software and 3D printing technology. We term this approach Innovation as Data. These two approaches differ substantially in terms of their views on the role of consumers, the relevance of firms, and the nature of products. In this article, we introduce the concept of Innovation as Data and discuss its relation to the digital revolution in general and 3D printing in particular. In addition, we offer a set of recommendations for managers seeking to leverage this new innovation approach and ideas for scholars interested in researching this intriguing domain. 94

108 Linking User Innovation and Social Innovation - An Explorative Case Study on Lead User Identification in the Humanitarian Sector Paper Daniel Kruse (Hamburg University of Technology), Moritz Göldner (Hamburg University of Technology), Cornelius Herstatt (Hamburg University of Technology) Despite comprehensive research on user innovations in a wide range of industries and regions, there is only little evidence for the existence of userinitiated solutions in the international humanitarian sector with its focus on social innovations. This is particularly striking as both, literature on social innovation and user innovation, show conceptual similarities. To fill this gap, our study is the first to apply the 'Lead User method' in this sector in order to identify user-driven social innovations that enhance the resilience of individuals or communities towards floods in rural and semi-urban areas in Indonesia. This study originated from a joint research project of the International Federation of Red Cross and Red Crescent Societies (IFRC), Geneva and Jakarta, the Indonesian Red Cross (Palang Merah Indonesia, PMI) and the Institute for Technology and Innovation Management (TIM), Hamburg University of Technology (TUHH) that was conducted between October 2016 and February By surfacing 25 heterogeneous innovations, we show that the Lead User method is an effective tool for identifying social innovations in the humanitarian sector in the area of flood resilience in Indonesia. Most of the solutions target flood risk prevention which has been an unresolved issue within the humanitarian sector so far. Our findings also confirm similarities between social innovation and user innovation, for instance, in terms of motivation and free revealing of solutions. Overall, our paper contributes to a challenge the humanitarian sector is facing since a considerable time: the need for localization of aid. By transferring a recognized method for innovation identification from the private to the humanitarian sector, we introduce a new path towards empowering local capacities for disaster risk reduction. 95

109 Universities as Lead Users Paper Sofia Patsali (Scuola Universitaria Superiore Pavia and BETA Strasbourg), Stefano Bianchini (BETA Strasbourg), Patrick Llerena (BETA Strasbourg) We explore the impact of a large French university on the innovative performance of their suppliers of research materials. By using a quasiexperimental empirical framework, we find that university suppliers have a higher propensity to introduce new-to-the-market product innovations compared to other firms belonging to the same sectors and with similar characteristics. In contrast, we find no significant impact on process innovation and on R&D intensity. Our results provide quantitative support to the hypothesis that universities may act as lead-users via their demand for specialized equipment and materials. 96

110 Customization in the Smart Product Age -Consumer s Response to Smart Products for Product Customization in the Usage Stage Poster Ning Wang (Aachen University), Frank T. Piller (Aachen University) In this paper, we test consumer s attitude towards the new concept for consumers to achieve customized and user-specific products in the smart product age, that is Product Customization in the Usage Stage (PCUS) with Smart Products (SPs). On the contrary to the existing concept of customizing products online during the point of sales, a new class of smart products- Adaptive customizable smart product (ACSP) (enabled by recent digital technologies and Internet of Things) allows consumers to configure them to achieve personalized product or service via embedded toolkits (e.g. connected apps) according to different usage context. The objective of this study is to investigate consumer s perception of ACSP by comparing two different SPs scenarios for PCUS SPs with autonomous personalization and SPs with user customization. Equipped with advanced ICT such as microchips, software, and sensors to collect, process, and produce data (Rijsdijk and Hultink 2009). SPs possess different degrees of smartness in terms of, e.g., autonomy, adaptability, multi-functionality or ability to cooperate (Rijsdijk and Hultink 2009). Inspired by the classification of adaptive system and adaptable system (Bunt 2004), we see two possible scenarios of ACSP. 1. Adaptive smart products (ASP) - SPs with autonomous personalization Sustained by smartness, such as adaptability and autonomy, this type of SPs can automatically adapt to the consumers and reconfigure themselves accordingly to a consumer's requirements without any dedicated user action. Thus, we call them Adaptive smart products (ASP). In this way, product smartness is used to save the entire product adaptation effort for consumers. However, consumers perceive loss of control and risk from high levels of product autonomy (Folkes 1988; Rijsdijk and Hultink 2009). Moreover, SPs lead to perceived complexity since they are adapting automatically like a black box and are difficult for users to understand (Rijswijk and Hultink 2009). Furthermore, the creative value from self-customization may get lost by the 97

111 automatic adaptation by a SP (Franke et al. 2010). 2. Customizable smart product (CSP) - SPs with user customization Some SPs which we call Customizable smart products (CSP), try to balance both benefits and disadvantages of automatic adaptation by integrating product customization possibility (in terms of, e.g., an embedded toolkit) into SP. CSP provide embedded toolkits or connected apps to facilitate user to get customized product or service in use. For example, Philips Hue smart light allows users to customize their preferred color effect and brightness for different situations like reading, romantic moments or parties etc. via a connected app. In this scenario, consumers are able to exert their influence over the product and also the customization process directly, so that they demonstrate their competence and mastery over the enjoyment, therefore the user can obtain the feeling of control (White 1959, Jiang et al, 2009). Based on control theory and literature from system adaptivity/adaptability and mass customization, we propose the research model as seen in Fig.1. H1-H4: CSP, in comparison with ASP, leads to higher perceived control, higher perceived enjoyment, higher perceived creative value and lower perceived complexity. H5- H8: Perceived control, perceived enjoyment, perceived creative value is positively related with intention to use; perceived complexity is negatively related with intention to use. To test the research model, we implemented an online survey which describes two above mentioned ACSP scenarios about smart sport shoe as following. ASP: ''Based on highly automatic and autonomous analyses, this smart sport shoe will set personalized running objectives for you. Via a user interface it will provide automated feedback about your running style and behaviour. The type of feedback (e.g. audio, vibration, display etc.), the frequency of the feedback and the content of feedback (e.g. virtual coaching advice, stance recommendation, running technique analysis etc.) are all predefined there is no user setup necessary.'' CSP: ''Based on user controlled and customizable analyses, this smart sport shoe will help you to reach your running objectives. Via a user interface it will also provide adjustable feedback about your running style and behaviour. You are able to control the way of feedback and interaction through this interface. For example, you can customize the type of feedback (e.g. audio, vibration, display etc.), the frequency of the feedback and the content of feedback (e.g. virtual coaching advice, stance recommendation, running technique analysis etc.) as you want. You can also turn the feedback function off.'' Now this survey is issued out and in the data 98

112 collection process. We plan to present our results, conclusion and research findings in the conference. Overall, our study contributes to research in the domain of user customization and co-creation by extending our existing understanding of product customization by introducing a new perspective of SPs and exploring the value of different kinds of ACSP to consumers. For practice, our results may be insightful for the design of SPs for PCUS and influencing consumer s adoption of ACSP for example intention to use or purchase. Keywords: Smart Products(SPs); Product Customization in the Usage Stage (PCUS); Adaptive Customizable Smart Product(ACSP); Perceived control; Intention to use. 99

113 The Prototype Used in the Implementation of the Need-Solution Pairs - The Prototype Use of Single-User and Multi-User to Relate Need-Solution Pairs Poster Akimitsu Hirota (Kindai University) This study is to examine the mechanism of pair joining process in Need-Solution Pairs. Need-Solution Pairs is a framework of a dynamic viewpoint that needs and solution are simultaneously created while mutually exploring. The feature of this research is to grasp the mechanism of Need-Solution Pairs from two viewpoints. First, pay attention to the role of the prototype. Second, pay attention to the effect of using the same prototype by multiple users. These investigations were conducted for the product development process of a car pedal called Naruse pedal. Naruse pedal is a pedal in which the accelerator and the brake are integrated. Therefore it is called "one pedal". There are few reports of user innovation cases of automobile products belonging to scrape type (Clark and Fujimoto 1991). In this research, we consider the promotion of Need-Solution Pairs by prototype based on the survey of Naruse pedal development process. 100

114 How Learning Effects Lead to Higher Quality and More Novel Solutions in Pyramiding Search Poster Nikolaus Franke (Vienna University of Economics & Business), Barbara Mehner (Vienna University of Economics & Business), Kathrin Reinsberger (Vienna University of Economics & Business) In this project we explore learning effects in the pyramiding search technique. This technique is one of the most important and frequently used ones for identifying rare valuable individuals like experts or lead users (e.g. Stockstrom et al., 2016). Identifying those rare individuals is essential for the ability of organizations to achieve truly new, disruptive innovations (e.g. von Hippel, 1988). This sequential method has been analyzed and compared to other, parallel search methods in past studies, but only regarding their efficiency (Stockstrom et al., 2016; von Hippel, Franke & Pruegl, 2009). Especially for the pyramiding search, recent research calls for further analysis regarding its application (Savino, Petruzzelli, & Albino, 2017). We follow first clues indicating that learning effects take place in the black-box of pyramiding search processes (e.g. Stockstrom et al., 2016, von Hippel et al., 2009; von Hippel et al., 1999). Due to the sequential, reference-based nature of this approach, learning processes could occur regarding continuous adaptations in e.g. the search goal definition (what exactly is the problem that needs to be solved?), the search field definition (where can I find good solutions to this?) and the problem framing (how to ask the right questions to trigger valuable solutions?) in a first step of problem solving (Treffinger, Isaksen & Stead-Dorval, 2006). These learnings could improve search outcomes regarding important aspects of innovation processes - namely idea quality and novelty (Dean et al., 2006) - compared to parallel search methods that can only provide solutions to the initially defined search parameters and consequently depend heavily on the quality of this initial definition. Therefore, we address the following research questions: I. What can be learned within the pyramiding search method? II. How do these learnings affect the outcomes of pyramiding search compared to a parallel search 101

115 regarding their quality and novelty? Parallel search approaches are associated with a limited solution space (by the predefined sample) and once the search process is started, the parallel nature of it allows no adjustments in the process. As a contrast, in a sequential process like pyramiding, learning between the nodes in the reference chain is possible (Loch et al., 2001; Thomke et al., 1998). Those potential learnings need to be explored regarding their value for innovations. Understanding these effects could yield to the systematical creation of improved innovation processes in the future. In combination with being more efficient than other methods (Stockstrom et al., 2016; von Hippel et al., 2009), this could give clear recommendations to firms seeking disruptive innovations to intensively deploy the pyramiding search method. The method is based on an extensive comparative analysis of the outcome and process of the pyramiding and a parallel search technique in an experimental setting. Participants will independently work on 50 real-life innovation problems either using the pyramiding or a parallel search technique. For both cases, the same conditions will be applied and during the course of the experiment, participants activities will be tracked and all conversations with external information sources in the search process will be documented. While we are currently in the conceptualization phase, by July 2017 we expect to have made significant progress regarding the study design. We aim to contribute to open and user innovation research in the following: One of the important questions is how to transfer external, innovation relevant knowledge to an organization that is looking for it (e.g. von Hippel, 1994). So far, pyramiding as a suitable strategy usually has been examined in case studies (e.g. Hyysalo et al. 2015; von Hippel et al., 1999) or with simulations (e.g. Stockstrom et al. 2016; von Hippel et al., 2009). These methods show the structure of reference chains and networks (e.g. Kratzer, Lettl, Franke & Gloor, 2016) but not what happens between the nodes in the search chain (Stockstrom et al., 2016) and how the process develops for a given topic directly compared to parallel search processes. We explore the underlying mechanisms with our experimental approach. With our expected results we propose pyramiding as the long sought tool to systematically trigger unexpected new information and valuable recombination of existing knowledge elements (Savino et al., 2017). This combination of existing knowledge elements is connected to the phenomenon of serendipity, where people discover valuable information by chance (Merton, 1945). As serendipity can play an important role 102

116 in innovation (e.g. Hargrave-Thomas, Yu, & Reynisson, 2012), it has received a lot of attention in search literature and many attempts have been directed at consciously enhancing it in specific contexts (e.g. Eagle & Pentland, 2004). We hope to identify mechanisms of pyramiding that can create serendipitous outcomes. Additionally, we will extend problem solving literature in a valuable way. The entire pyramiding search chain can be seen as problem solving process, where each of the steps are repeated several times. The process of creative problem solving that consists of understanding the problem, generating ideas and preparing for action, i.e. developing solutions (Treffinger et al., 2006) can take place even at each individual node in the pyramiding search chain and results can be directly integrated. Recent research on problem and solution forming claims, that in contrast to the traditional concept of problem solving, where only after a clear problem definition the solutions are developed (Duffy, 2003), problems and solutions can develop simultaneously or in a reverse order (von Hippel & von Krogh, 2015). By exploring the development of problems and solutions in an empirical setting, we anticipate to contribute to settling this controversial debate. 103

117 The Management of User Driven Innovation Processes: Towards a Systematic Review Poster Guido Bortoluzzi (University of Trieste), Khatereh Ghasemzadeh (University of Udine) In traditional point of view, marketing methods of collaborating with users considered the primary needs and interests of users (Kristensson, Gustafsson, & Archer, 2004), but users can be the sources of innovation, not just assistants to produce new products and services (von Hippel, 1986). User driven innovation (UDI) has become an interesting topic and achieved considerable attention in innovation studies and practices after 2000 (Hyysalo, Repo, Timonen, Hakkarainen, & Heiskanen, 2016). A firm should shift from producer-centered innovation toward customer-driven innovation in order to enable continual, sustainable innovation (Desouza et al., 2008). A great deal of conceptual and empirical studies has been focused on ways and toolkits (von Hippel & Katz, 2002) and management of user communities (Fuller, Matzler, & Hoppe, 2008) in order to provide co-creation enabling users to experiment and innovate (Jeppesen & Frederiksen, 2006). Also, many studies have deepened innovation approaches based on the involvement of specific categories of users, such as in the case of lead-users (Luthje, 2004; von Hippel, 1986). The aim of this paper is to conduct a systematic literature review in the scientific domain of user-driven innovation studies to provide a state-of-the-art picture of the UDI phenomenon, with particular attention to the organizational routines to organize it and to the managerial processes to manage it within the firm. After selecting and reviewing some of the most cited articles in the UDI area and after reading already available reviews on this topic (Bogers, Afuah, & Bastian, 2010; Greer & Lei, 2012), it seems to us that the literature is providing abundant attention on the preconditions and on the consequences of the process of users involvement while overlooking some phases of the process itself. For example, we know quite well what are the strategic antecedents - such as, having an open innovation strategy - that create the optimal preconditions for UDI to settle in a firm. On 104

118 the other hand, the literature has clarified in a quite exhaustive way which kind of users (individuals vs. communities, normal vs. lead users) should be involved in different innovation contexts and phases. What, in our opinion, we miss a better picture of the organizational antecedent, the organizational routines, and the managerial processes through the which UDI-oriented interactions and relations are managed within firms. Even though the research stream of UDI has been developing over a period of almost four decades, our impression is that just a minority of studies within this wide literature have been focusing on the organizational design and the managerial processes of UDI processes. This is, in our opinion, a relevant topic since good organizational and managerial processes are at the base of a fruitful exploitation of UDI strategies and practices (Keinz, Hienerth, & Lettl, 2012). To carry out this review, we relied on Web of Science that was searched using various combinations of 14 different but related keywords such as user-driven, customer driven, user involvement, customer cocreation, innovation, co-development, new product development, etc. Only articles published articles in scientific journals were considered, while book chapters and conference papers were not included. Further papers were gathered from the bibliographies of the most recent papers related to UDI. The total number of entries using keywords and bibliography search was 340. After retrieving the papers, and the bibliographic information, data were exported to an Excel table. The following phase was an initial review of titles, journals, and abstracts in order to exclude completely unrelated papers. In the first filtering process, 28 papers were excluded because they were not related to organizational and managerial issues. Another primary exclusion factor was eliminating papers dealing with open innovation and innovation in general with no explicit reference to UDI. The initial process narrowed down the sample to 312 articles. In the next step, the remaining articles were graded from 1 to 5 in order to determine how close each article was to the UDI topic where 1 refers to least related papers to UDI and 5 refers to the highest closeness. The results of coding have been double-checked: the author of this review and her scientific supervisor coded independently the 312 articles and existing differences in the evaluations were reconciled. Articles not reaching the 3 threshold for both the evaluators were excluded from the review. In this level, more precise exclusion criteria considered to achieve the most promising articles in UDI topic. The most important exclusion criteria were: 1) papers focusing on innovation practices, 105

119 not UDI 2) papers grounded totally on open innovation theoretical framework 3) dealing with user experience and not with direct involvement of users 4) papers purely related to computer science and healthcare without contribution to UDI 5) buyer-supplier collaboration for new product development. In total, 270 papers remained and were included in the following phases. Currently, I am working on categorizing the papers according to the following classifications: a) type of paper b) theoretical framework c) methodology and method d) main findings In the next phase, all the articles will be further classified using a more fine-grained classification that will be based on emerging topics. Thus, no results can be shared at this stage, but some preliminary results will be available in the final version of the poster that will be presented during the conference. Guidance and critical feedback of the community attending the conference will help me in providing a comprehensive and structured systematic review of this fast-growing literature, and in particular on the organizational routines and the managerial processes that trigger and allow the collaboration process. Furthermore, the exact proofread of the poster will be accomplished until the time of the conference and the possible writing mistakes will be corrected and the quality will be enhanced. 106

120 User Knowledge Utilization and Distributed Sensemaking Poster Andreas Benker (Aalto University) Knowledge about users and their needs has been identified as valuable source for firms to innovate. However, it is challenging for firms to capture user knowledge that lies outside their organizational boundaries because it is difficult to identify suitable users and understand their ideas and needs. Moreover, even though user-generated ideas might be original and reveal promising value, they might not be feasible to implement given the lack of knowledge users have about the firms processes compared to professionals working in an organizational context. Users possess knowledge about innovative ideas that would provide value to a company whereas employees possess knowledge about the feasibility of such ideas given that they know the firms innovation processes and structures. One possibility for firms to overcome the difficulty of obtaining and utilizing external user knowledge is to incorporate users into their organization. Knowledge about user needs can be obtained through employees who are either also lead users of the firm s product or have been professional users in their previous working environment. These embedded users have been subject to discussion regarding the different roles they occupy within a business unit as user representatives, networkers, idea promoters, and change agents as well as the capabilities they deploy during ideation, development, and marketing phases of the innovation process. However, it remains unclear how they make user knowledge comprehensible to other R&D functions and translate it into the development process accordingly. In order to improve our understanding of user knowledge utilization in innovation processes, I carried out an ethnographic study of product development in a medical device manufacturer s R&D department as participant observer. During the three years of fieldwork I was at the site three to four days a week following one project throughout the development phases from concept development to marketability. Understanding knowledge creation in a social context requires more than solely verbal accounts of informants, but rather how it is reflected in people s attitudes and behaviors. Conducting participant observation over a 107

121 period of time enabled me to converse with the people being studied in their natural setting and better understand their official and unofficial practices, norms, and values. The findings highlight that translating user knowledge into an innovation process implies a sensemanking process by the respective R&D functions involved. Elaborating on distributed sensemaking, this study shows how embedded users support the understanding of user knowledge on a team level across different R&D functions. Embedded users contribute actively to conceptualizing user knowledge by articulating and validating user requirements. Furthermore, their evaluative function during design iterations enables re-contextualizing design implementations from the user perspective. Therefore, embedded users facilitate the collective induction of new meanings regarding user knowledge in the innovation context. 108

122 When the Use of Open and User Innovation Methods is effective: The Moderating Effect of the Organizational Setup Poster Michael Nobis (Vienna University of Economics & Business), Nikolaus Franke (Vienna University of Economics & Business), Peter Keinz (Vienna University of Economics & Business) At the heart of research on open and user innovation lies the assumption that extensive utilization of open and user innovation methods and instruments increases companies success in innovation efforts (Chesbrough, 2003; Christensen, 1997; Von Hippel, 2005; Lilien et al., 2002). Nevertheless recent studies show that this positive relationship is oftentimes found to be unstable, weak or in specific situations even reversed (Laursen and Salter, 2006; 2014; Foss et al., 2011; 2013). Studies investigating this phenomenon have found companies organizational practices especially with regards to their internal communication, incentive models and delegation of decision making to play a decisive moderating role on the relationship between usage of open and user innovation methods and companies innovation success (Foss, 2011). An ineffective organizational setup, therefore, seems to be a potential explanation for the contradicting results in studies on open and user innovation. Even though the practical relevance of research in this area is extensive, up to now major studies have only investigated singular processes or individual parts of the organizational setup (Foss et al., 2011; 2013). No study, however, has analysed the effects of a comprehensive set of organizational setup elements (i.e. organisational structure, corporate climate, innovation process setup and control, coordination & incentive mechanisms). As a result the central research question in our research is: Is organizational setup a moderator on the influence of open and user innovation methods on companies innovation success? In order to answer the research question we are conducting a large-scale online survey among the top-management of 4000 companies in Germany from May- June The questionnaire being used for the survey covers a broad range of questions on innovation success (OECD 2005) as well as questions on all relevant 109

123 elements of the organizational setup according to the framework presented by Burton and Obel in 2003 (i.e. innovation strategy, role of top-management, organisational structure, corporate climate, innovation process portfolio/setup and control, coordination and incentive mechanisms) (Burton & Obel 2003). The sample for the survey is randomized and covers all industries, company sizes (up to 5000 employees) and ages. By July 2017 the survey will be completed and initial findings can be presented at OUI 2017 Expected contribution from our research is an improved understanding of factors currently limiting the effectiveness of open and user innovation instruments in companies. We thereby extend existing literature on open and user innovation and provide an explanation for the phenomenon of an unclear relationship between the usage of open and user innovation methods and instruments and companies innovation success. For practitioners we deliver key insights on how to set up and optimize their companies organization and internal processes to substantially increase success of innovation activities. 110

124 Law and IP 111

125 The Strategic Coupling of Intellectual Property Management and Co- Creation Paper Anja Tekic (Skolkovo Institute of Science and Technology), Kelvin Willoughby (Skolkovo Institute of Science and Technology) We argue that success of co-creation in product innovation projects will generally require adoption of an intellectual property management approach that is appropriately matched to the particular type of co-creation practiced by a company. To answer the question of how intellectual property management may be coupled with co-creation in product innovation projects we developed a two-dimensional conceptual framework with intellectual property management approach (open and closed) on one axis and co-creation type (company-to-one and company-to-many) on the other axis which enabled us to postulate four distinct intellectual property management strategies in cocreation, namely the Compensation-Retention, Appropriation-Compensation, Benefaction-Benefaction, and Appropriation-Benefaction/Compensation strategies. Our analysis of a carefully selected set of co-creation cases confirmed the value of this conceptual framework, revealing that all postulated generic strategies are indeed employed by companies, although the balance of preferences for each strategy was different from what we expected based on presumptions frequently expressed in the literature. 112

126 The Interplay between User Innovation, the Patent System and Product Liability Laws: Policy Implications Paper Stijepko Tokic (Northeastern Illinois University) The enclosed article seeks to broaden an on-going policy debate about user innovation, which refers to individual inventors or firms that expect to benefit from using (rather than selling) their overwhelmingly unpatented innovations. In particular, the article examines the interplay between user innovation, the patent system and product liability laws. The influential empirical studies of MIT s Eric von Hippel and his colleagues have documented the extent and importance of user innovation and, subsequently, have prompted discussions about reevaluating laws and policy related to the patent system. It has even been implied that the evidence of thriving user innovation indicates that innovation can prosper without the patent system altogether, as user innovators do not need the external incentives to innovate provided by the patent system. This article challenges that notion by explaining the multifaceted role of the patent system in promoting innovation. Moreover, the article argues that the lack of exposure to product liability laws offers an explanation as to why user innovation is thriving, since user innovation is principally outside the scope of the product liability laws, so user innovators can take more risks and be more innovative than producers/manufacturers. The article identifies diffusion as the main issue in user innovation, and argues that diffusion of user innovation is currently not substantial enough to promote innovation, but is significant enough to raise product liability issues, which have largely been ignored in the user innovation literature. The article concludes by warning that calls to reevaluate intellectual property rights fueled by research on user innovation should be taken with much caution, not only because of the barefaced issues related to promoting and diffusing innovation, but also because of latent issues related to product liability laws. 113

127 Users, Patents and Innovation Policy Paper Katherine Strandburg (New York University School of Law) A patent-free user innovation paradigm is likely to be successful and socially desirable when an invention s value to users has a substantial non-competitive component. If a user innovator values an invention primarily for providing a competitive edge, the patent-free user innovation paradigm is not viable. Most such inventions have little social value. Some, however, such as improved manufacturing processes, produce significant collateral value for non-users and should be encouraged. Patents may be important for these user innovations. Socially beneficial policy interventions to buttress the patent-free user innovation paradigm might include tools and infrastructure to support user communities and changes to patent doctrine, such as accounting for user innovation in assessing nonobviousness, patentable subject matter exemptions, particularly for many types of processes, and user exemptions from infringement liability. 114

128 Open-Source and For-Profit Innovation in the World of Things. A Pilot Study in the Circular Economy in Flanders (Belgium) Poster Geertrui Van Overwalle (University of Leuven), Lodewijk Van Dycke (University of Leuven) Over the past years, innovators who collaborate with others have received increasing attention from economic and legal scholars. First of all, there is the literature on open innovation (Chesbrough 2003, 2006), paying attention to enterprises who use intellectual property (IP) rights as bargaining chips in collaborative research and development contracts. Second, there is the literature on user innovation (von Hippel 2005, 2016), paying attention to the role of users in general, and of lead users in particular for a fruitful corporate innovation strategy. Third, there is the literature on open source software (Stallman 2002) and information products (Benkler 2006). Recently, the open source trend has spilled over to the world of tangible, physical products: some innovators of physical products explicitly want their innovations to remain open for all. This open source approach in the world of physical products, or the world of things, is epitomized by initiatives such as open hardware (Söderberg 2013), open design (Raasch, Herstatt, and Balka 2009) and open source appropriate technology (Pearce 2012). In the literature on user innovation and on open source innovation little attention has been paid so far on the attitude of innovators of physical products towards IP law, especially towards patent law. It has been stressed in legal scholarship (Elkin-Koren 2005; Dusollier 2007) that safeguarding the open source ideal for software and creative works is guaranteed via copyright complemented with some kind of copyleft licence regime. It has been argued that acquiring a patent may well be necessary to establish and enforce open source principles with regard to physical products (Van Overwalle 2015). Scholarly work on the IP attitude of innovators of physical products who wish to adhere to the open source norm, and who are motivated by a for profit motive at the same time, is even more scarce. Innovators having the ambition to fully remain open source, and at the same time, to found a 115

129 viable enterprise through their innovations, have received little or no attention in legal scholarship. A concrete example of this open-source for-profit trend is the Brussels bio-engineer, Elise Elsacker and her company Magma Nova: Elise has a passion for bio-fabrication and truly believes in the power of a collaborative society, instead of a heavy bank account and 100 patents. She wants to establish Magma Nova as an open source, albeit liveable start-up from which she hopes to derive an income. In the project we have launched in the winter of 2016, it is this rare and rather unseen combination of open-source and for-profit with regard to physical production we want to explore. We wish to understand whether IP law in general, and patent law in particular, plays a role in the open source strategy and the business plan of open source innovators of physical products. The project studies the innovation and sharing behaviour of innovators fitting the following profile: innovators (1) aiming to achieve significant innovation, (2) playing a role in the production of physical things, (3) adhering to the principles of open source, and (4) aspiring to make a profit out of entrepreneurial activity. The pilot study focuses on innovators fitting the four criteria above and being active in the circular economy in Flanders. We have put up a database of all open source innovators of physical products in the Flemish circular economy through Vlaanderen Circulair (the competent government department), Generation-T (a network of young leaders towards a sustainable society), the Shift (a network of sustainable companies which are active in Flanders), MakeSense (a global community mobilizing volunteers all around the world to help social entrepreneurs solve their challenges ) and Hackistan (a community of hackers who aim to provide smart solutions). The database consists of about 20 entrepreneurs, whom we sent an to invite them for an in-depth interview. Four entrepreneurs (RegenerativeDesign, Yuma, Faro360 and Reagent) and Vlaanderen Circulair, the Shift and Hackistan agreed so far to meet for an interview. All interviewees were asked for the names of other innovators that fit the profile (snowball sampling). Interviews. The innovators that fit the profile and that responded to the were interviewed at length. The interview was semi-structured. The interview consisted of 22 main questions, which are loosely based on the analytical framework of Frischmann, Madison and Strandburg (2014). By now, all interviews are transcribed, and the coding has begun. The pilot study is a qualitative study. The study has many tenets of a grounded theory study, but 116

130 also involves many elements of a thematic analysis. However, the pilot study also contains elements which are not related to grounded theory at all. These elements are inspired by the fact that the study relates to a legal phenomenon studied by legal scholars. First and foremost, the pilot study attempts to provide an adequate description and explanation of ongoing practices. Second, the study aims to analyse the documented practices through the lens of the commons theory (Oström 2009 ), as exemplified in the area of knowledge production (Frischmann, Madison and Strandburg 2014). The knowledge commons framework enumerates and qualifies the different aspects of pools of knowledge : the background IP environment, the resources put into the knowledge commons, the community behind the knowledge commons, the rules for using the knowledge commons, the actors involved in the knowledge commons, the actual interactions, the generated outcomes, and, the relationships between these aspects. The poster submitted for the OUI conference reports on the pilot study which was conducted in spring During the OUI conference, we will present the first findings of the coding of the interviews. 117

131 Innovation Policy 118

132 Policy Choices in Supporting Collective Innovation: A Study of Pharmaceutical R&D Consortia Paper Joel West (Keck Graduate Institute), Paul Olk (Daniels College of Business) While many R&D consortia of the late 20th century focused on improving national competitiveness and other forms of advantage for consortium members, the rise of industrial open source cooperation in the early 21st century brought a new, more open model of R&D consortia that allowed open spillovers to nonmembers and society as a whole. This study utilizes a new database of 399 such consortia in the life science industries most organized as public/private partnerships funded by government and commercial interests. Using this data, we identify six distinct institutional forms three public led, three private led for organizing such cooperation in the biomedical sector. We compare and contrast these forms as each seeks to find its institutional niche, and more generally contrast the policy choices between public and private leadership, as well as different forms of publicly-led cooperation. 119

133 Identification and Classification of User Innovation: Implications for Future Research Paper Jakob J. Korbel (Technical University Berlin) An increasing number of national user innovation surveys provide evidence for a remarkable number of innovating users (von Hippel, 2017). A re-estimation of those results even imply an underestimation of this innovation phenomenon due to methodological reasons (Franke et al., 2016). We suggest that the applied user innovation identification and classification frameworks might pose another source of an imprecise assessment of user innovation. The applied frameworks in the literature are similar in many respects, but they differ in terms of the initial inclusion criterion, the applied commercial market availability criterion and the evaluation of the novelty value of user creations. With our analysis, we could show that depending on the selected criteria, the rate of user innovators can either decrease, for example, from 37,3 % to 9,8%, or increase by 12,7%. In order to detect underlying problems in the identification and classification of user innovations, we analyze the DIY Maker community Instructables.com, using a non-participant netnographic approach with computer-assisted data collection. The qualitative analysis of the community reveals two main factors that might explain the disparities in the publications: The existence of a parallel market and the non-consideration of the context and complexity of products. Based on these findings, we propose a uniform user innovation identification and classification framework. The implications for national user innovation surveys are both encouraging and daunting. On the one hand, the findings suggest that in surveys the number of user innovators is underestimated, as for example the application of the market availability criterion can lead to the exclusion of a multitude of potential user innovations prior to the researchers assessment. On the other hand, the strict application of the proposed definition of user innovation in terms of public availability would decrease the overall number of user innovators, for example, in the US from 16 million to 2,88 million. 120

134 Trajectories of Local Open Government: Investigating Managerial Perception of Innovativeness Paper Lisa Schmidthuber (Johannes Kepler University Linz), Dennis Hilgers ((Johannes Kepler University Linz) Over the past few years, there has been a substantial change towards openness in government organizations. Open government is associated with a new form of collaborating with citizens and an innovative way of organizing public service delivery. This opening up of administrative structures to external environment requires change within public organizations. To shed light on the new functions of public organizations, this study aims at investigating the factors that influence open government adoption at the local level. It is assumed that municipal level of open government adoption is influenced by organizational characteristics, manager characteristics, and characteristics of open government practices. We tested our hypotheses by conducting an online survey among city managers of Austrian municipalities and found significant empirical support for a positive influence of task knowledge, resource availability as well as benefit of open government as perceived by the city manager on municipal level of open government adoption. The study s findings contribute to research and practice. We (1) provide a model of open government adoption by drawing on the theory of innovation adoption, (2) empirically assessed innovation adoption of local governments and its antecedents, (3) point to characteristics of public organizations that directly influence open government adoption. 121

135 Innovation in the Household Sector: Definitions, Statistical Measurement and Policy Paper Fred Gault (UNU-MERIT Maastricht and Tshwane University of Technology This paper examines generalised definitions for measuring innovation in the household and other economic sectors. The long-term goal is the establishment of official statistics on innovation that go beyond those well established in the business sector. A systems approach is used to describe the actors, their activities and interactions in order to show that innovation is not an isolated phenomenon and that all economic sectors play a role. Policies to promote household sector innovation are considered along with the economic and social outcomes. 122

136 Open Social Innovation Dynamics and Impact: Exploratory Study of a Fab Lab Network Paper Thierry Rayna Novancia (Business School Paris), Ludmila Striukova (University College London) The aim of this research is to explore the dynamics and impact of open social innovation, within the context of fab labs and makerspaces. Using an exploratory methodology based on 11 semi-structured interviews of fab lab founders belonging to the Centres for Youth Innovation and Creativity (CYIC) programme a network of 170 fab labs located in Eastern Europe this research explores the impact of an adopting an open approach in relation to the different stages of social innovation (prompts, proposals, prototypes, sustaining, scaling and diffusion, systemic change) as well as social impact. The main results of this study are that while the CYIC programme provided each fab lab with similar initial conditions (identical funding, objectives and rules), the open social innovation approached adopted enabled to give birth to a wide diversity of fab labs, each being very well adapted to the local environment, social needs and constraints, and able to deliver social impact in just a matter of years; a result that would be hard to achieve with a centralised top-down approach. Furthermore, this research identifies the main difficulties social entrepreneurs encounter as a part of the open social innovation process, as well as means to overcome them. In this respect, this study adds to the literature on fab labs by providing a more comprehensive view of the challenges faced by fab labs (and makerspaces) founders, as well as suggestions of strategies enabling to ensure their long-term sustainability. 123

137 Translating User Innovation Research to Innovation Policy Action Poster Jari Kuusisto (University of Vaasa), Liting Liang (University of Vaasa) Translating user innovation research to innovation policy action Dr. Jari Kuusisto, University of Vaasa Dr. Liting Liang, University of Vaasa June, 2017 Extended Abstract Background Several drivers and barriers influence how effectively academic research results can be translated into evidence-based public policies and interventions that increase social welfare. Interest towards this translation process is growing since there is an increasing pressure to improve the societal impact of research and public policies alike. Universities are seen as well-informed, relatively objective advisors that can create social welfare e.g. by contributing evidence-based policies (Mirvis, 2009, Praznik, 1999). Also tight government research budgets are driving universities to demonstrate measurable impact of their research on public policies, and social welfare. Also the major research policy programmes, such as Horizon 2020, manifest the fact that an increasing number of stakeholders are showing interest towards societal impacts of higher education (EU Commission, 2011). Despite these strong drivers, the translation of research findings into policy action often seems to be a major challenge for all parties involved in the process. There are three groups of actors that are key to the translation process, including academics/researchers, policy entrepreneurs and policy implementing agencies. Most academics/researchers would find it rewarding if policy makers adopt their idea into a policy. Other benefit for researchers is that collaboration with policy makers and agencies may create an opportunities to collect new and unique data for the analysis. However, academics need to publish in academic journals and books with academic/technical writing style. The language presenting research outcomes is often incomprehensible to the policy makers. Academic credentials alone do not make scientists ideas translatable into policy action. Policy entrepreneurs, who are looking for new ideas are typically bound by strategies and policy priorities that direct their attention towards certain types of new ideas. Further more, policy entrepreneurs have limited time to study and familiarise themselves with complex academic projects. Hence, new ideas need to be communicated 124

138 effectively to policy makers who demand policy brief /summaries, concrete reports, guidance, documentation and evidence-based finding with clear, convincing, interesting and short language to be acceptable. However, the value of such non-academic outputs have not been widely recognised by academics. Furthermore, most academics do not have knowledge on policy makers interests at that particular moment, and areas where they seek to develop new policies. Neither understanding of feasibility for turning their research based ideas into policy, or tools and means that policy makers have at their use. Research questions & methodology Although universities play an important role in offering research evidence for policy making, their strategic approaches to influencing policies and decision-making are often very weak. We aim to answer the following question: how to foster the translation of user innovation research findings to policy action? We attempt to answer the question by exploring the antecedents for translation of research findings to policy action. The key here relies on understanding the interaction and communication between these different groups of key actors during the process. Knowledge management, knowledge sharing and absorptive capacity are also among the key concepts of the framework to be developed and used in the empirical analysis. We seek to address these research questions by analysing policy development case studies. They can provide real life insights and evidence on the interaction between policy entrepreneurs and researchers. Finnish Demand- and User-driven innovation policy development reports translation of research results over four years time period. It was a joint effort by the Ministry of Employment and the economy of Finland, and an international research team lead by Eric von Hippel from the MIT (Ministry of Employment and the economy, 2010). As a research topic user innovation has been well established over the last 20 years time and it has much potential to contribute innovation policy. Much of the credit belongs to growing user innovation research community and its leading researchers who have systematically developed theoretical frameworks such as Free Innovation Paradigm (see e.g. Hippel, 2017). It has significant implications for innovation policymaking in several areas such as intellectual property, innovation measurement and legal rights of individuals to develop and use innovations (von Hippel and Jin, 2009; Torrance and von Hippel, 2015, von Hippel 2017). Despite of the apparent policy implication of user innovation research, surprisingly little is known about issues related to the process of translating user innovation 125

139 research to innovation policy action. Slow pace of research turning into policy action is evident despite the fact that interest towards user innovation has surged also as a result of enabling technologies that have given a boost for the phenomena. For instance, the surge of digital tools has enabled very low cost pooling of user innovator knowledge, free sharing of ideas and resources as well as diffusion of user innovations (Baldwin and von Hippel, 2011; de Jong et al., 2015). The second policy development case reports transnational work on service innovation policy over a seven years period conducted by the OECD, EU and a number of ministries and policy agencies from several countries. This international study will add up to the findings of User innovation study. In both cases empirical material consists of participant observation, meeting documents, presentation materials and official policy documents. Expected outcomes We attempt to develop and examine the conceptual framework based on the policy cases related to user innovation and service innovation. This framework seeks to explain the key antecedents for the effectiveness of research translation. We seek to understand the role of interaction, communication and absorptive capacity in the translation process. Research translation is a complex, dynamic and non-linear process and better understanding of it can yield benefits for science as well as the society as whole. Improvement of this process is equally important for the universities and policy makers. 126

140 Problem Solving and Toolkits 127

141 Selective Broadcast Search Improving Problem-Solving s Efficiency Using a Company s Internal Crowd Paper Christian Pescher (FAU Erlangen-Nürnberg), Katja Hutter (University of Salzburg), Johann Füller (Hyve AG), Michael Heiss (Siemens AG) While broadcast search is quite popular to attract external solvers, companies rarely apply it to crowd source problems to their workforce. An analysis of a unique dataset from a multinational company s internal problem-solving community with 1,039 broadcasted problems over four years reveals that broadcast search can be successfully implemented within a company s workforce. Sending out each problem to every potential solver uses employees paid working hours inefficiently. To improve efficien-cy, the responsible managers introduce a selective broadcasting approach, which uses a) the employees digital footprint to distribute each problem to those employees who are likely to solve it and b) the prob-lem s potential impact to determine the number of recipients. Despite a significantly lower number of invitations, selective broadcast search substantially reduces the resources each solver invests, but only slightly affects the response number and the solving probability in the first year, both of which even increase in the second year. Surprisingly, problems are more likely to be solved if they are a) more com-plex and b) framed so that everyone can easily understand them. The problem owner s prior effort in the community (reciprocity) is more important than the formal position within the company (signaling). These findings contribute to the emerging literature of open and distributed problem solving by study-ing the problem and people characteristics that lead to successful problem solving. 128

142 Value Development During the Self-Design Process: A Demonstration and Explanation of the Swoosh Effect Paper Nikolaus Franke (Vienna University of Economics & Business), Franziska Metz (Vienna University of Economics & Business), Page Moreau (University of Wisconsin) We ask two research questions: (1) Does the value development during selfdesigning with a configurator follow a power law or a swoosh -curve? The relevance of this question results from a puzzle: While research repeatedly found a strongly increased willingness to pay for self-designed products than their off-the-rack counterparts, the market share of self-designed products in any given category remains low and the rates of customers who do not finish the process are extremely high. It is thus likely that the subjective value consumer expect to derive from self-designing develops during the process of self-designing. But how? (2) In which stage of the self-design process does which category of knowledge influence value? It is plausible that the subjective value the customers expect during self-designing is a function of their current knowledge. We argue that the influence of different categories of knowledge (knowledge how the configurator works, knowledge about one s own preferences) changes as the process unfolds. Understanding the drivers of value is not just of theoretical import, it is also the basis for managerial implications. We gathered data in a laboratory experiment where 508 participants selfdesigned Nike Sneakers. In order to measure subjective value in an unobtrusive, data rich, longitudinal within-subject manner, we employed FaceReader technology. Facial expressions, most often uncontrolled and spontaneous, are one of the primary channels for humans to transmit emotional signals. In order to validate our dependent variable, we used BDM auctions at randomly predetermined points in time during the self-design process (between-subject measurement). The two categories of knowledge were also measured during this interruption using survey technique. For our analyses, we employed OLS regressions and multilevel fixed effects regressions. By introducing a truly 129

143 longitudinal perspective and shedding light on the development of value in selfdesigning, our research contributes to the literature on value elicited by toolkits and its driving forces. Our research offers a solution to the puzzle sketched above: Despite their great potential, self-designed products have not yet captured greater market share because extant configurators are typically not adapted to the learning processes customers undergo. Our core finding is that the value development during self-designing follows a swoosh -curve. An initial high value expectations is followed by a rapid decrease, caused by the necessity to learn about the toolkit functioning (in the first phase) and one s own preferences (in the second phase). The resulting minimum level may be the exact moment when potential customers abandon the configurator in frustration unaware of the typical sharp increase of value they would have achieved directly after this valley of death. 130

144 Problem Solving Without Problem Formulation: Documenting Need- Solution Pairs in a Laboratory Setting Paper Ruth M. Stock (Technische Universität Darmstadt), Eric von Hippel (MIT Sloan School of Management), Christian Holthaus (Technische Universität Darmstadt), Nils Lennart Gillert (Technische Universität Darmstadt) It has been hypothesized by von Hippel and von Krogh (2016) that problem solving often occurs via the simultaneous recognition of both a need and a responsive solution without prior formulation of a problem being required. If this hypothesis is correct, significant new opportunities are opened up for both research and practice. The absence of a requirement for problem formulation can significantly reduce the effort and complexity of problem-solving. It also eliminates constraints on the range of possible solutions that a problem statement inevitably imposes, and so may enable the discovery of more creative, novel, and valuable solutions. In this paper, we conduct a first test of the von Hippel and von Krogh hypothesis via a laboratory experiment. We find our subjects do frequently engage in need-solution pair problem solving even when not requested to, and that a high fraction of the solutions they report follow the need-solution pair (NSP) pattern. In Condition 1 in the experiment where no problem or solving activity was suggested, 73% of the solutions were of the need-solution pair problemsolving pattern; in the progressively tighter Conditions 2 (broad solving task specified) and 3 (narrow solving task specified), the frequency of NSPs totaled 56% and 37% respectively: the hypothesis is confirmed. We also find that both the novelty and creativity of solutions discovered via need-solution pair recognition are significantly higher than solutions discovered via the traditionally assumed need-first pattern, and that their general value and utility are as good or better. On this basis, we suggest that need-solution pair problem solving is likely to be a phenomenon of high practical value as well as of high theoretical interest. 131

145 Scenario Design: Using Design Thinking in Scenario Technique Among Technology Clusters Paper Reimo Jahn (Helmut Schmidt University), Hans Koller (Helmut Schmidt University), Michael Andreas Zeng (Helmut Schmidt University) Scenarios are a widespread method for firms to cope with the erratic future. Despite its contribution for building future knowledge and informing strategic decisions, scenario technique is often confined by political discourses, autochthonous mind-sets, or a dominant orientation towards present or past events (Berkhout et al. 2014, Kunseler et al. 2015). Design Thinking can help to overcome these flaws, e.g., by encouraging creative interaction (with symbolic or tangible artefacts) and giving fresh momentum to tackle problems from unfamiliar perspectives (Shamiyeh 2016, Selin et al. 2015). Therefore, we regard Design Thinking methods as complementary to scenario technique in order to enhance its versatility, malleability, and innovative potential. Even though Design Thinking s practical potential has been outlined copiously (Johansson- Sköldberg et al. 2013) mostly for product and service design (Verganti 2008, Carlgren et al. 2016) only a small number of publications dedicates to the connections between design and scenarios (e.g., Hall et al. 2013, Chermack & Coons 2015). In this paper, we build on these primarily conceptual reflections and carve out intersections between design and scenarios in order to evince lacunae in scenario technique as starting points for Design Thinking s contribution. Moreover, we propose a collaborative workshop framework ( Scenario Design ) and report on first insights we gained from in-field experience with company representatives in three technology clusters (Renewable Energy, Aviation, and Life Science). We chose a multiple case study approach (Yin 2014) in order to explore how Design Thinking methods can be translated into collaborative scenario practice. We pre-tested our approach in four pilot workshops with engineering and management students. Later reflection on our practical experience helped to adapt the workshop procedure to our participants needs rather than adjusting it to theoretic rigidity. Data 132

146 collection during workshop interaction was based on observations, semistructured interviews, and self-administered questionnaires. Our insights indicate 13 intersections between Design Thinking and scenario technique. We also show that applying Design Thinking methods extends scenario technique in six different ways. Intuition-focused methods from Design Thinking allow for scenarios less biased by group think, political behaviour, or cognitive inertia due to present or past phenomena, and stimulate hidden creativity by embodying abstract future visions and re-perceive scenarios from unfamiliar perspectives. Therefore, our study contributes not only to the design literature by proving that Design Thinking is a powerful approach to improve the creation of future realities. It also adds to the foresight strand with empirical insights from designadded scenario technique. For practitioners we forward a pragmatic, creativitydriven, and flexible way of designing future scenarios in a collaborative manner. 133

147 Gaining Insight for Innovation Paper Peter Hu (University of Chicago), Shannon Heald (University of Chicago), Peter Malonis (University of Chicago), Howard Nusbaum (University of Chicago) One possible path to innovation is through insight. Insight has been defined as a sudden comprehension that solves a problem, reinterprets a situation, explains a statement, or resolves an ambiguous percept (Kounios & Beeman, 2009). While this definition offers a description for the physical experience of insight, it tells us little about the mechanisms that underlie the achievement of insight. Using several examples of insight in the perceptual domain and neural data, we argue that insight should be defined further as the sudden reorganization of attention towards appropriate problem-space (or solution-space) features through the application of apposite knowledge. With this new definition of insight we then explore how powerful insight can be in granting meaning. We further examine how such a definition of insight may bring us closer not only in identifying individuals who are the most insightful, but also in determining whether it is possible to conceive of insight as a skill that one can foster and develop. 134

148 Everyone is a designer Radical Innovation to Foster User Creativity in Toolkits Poster Evgeniia Filippova (Vienna University of Economics & Business), Nikolaus Franke (Vienna University of Economics & Business) Everybody is original. Everybody can design if not supremely, at least beautifully (Henry Wilson) Self-designing a product with a toolkit means, in essence, solving an ill-defined problem in a creative way (Franke, Keinz and Schreier 2008). The problem is ill-defined because consumers usually have only a rough idea about the desired product (Simonson 2005) and it is hardly possible for them to construct the product in all its details in their minds and just transfer the solution to the producer via a toolkit interface (Huffman and Kahn 1998). The process of designing a product via a toolkit is creative since the users want to design something novel (Jeppesen 2005; Franke, Schreier and Kaiser 2010). From this perspective, a good toolkit is a toolkit that supports customers in their creative problem-solving process (Franke and Hader 2014). However, the major part of mass customization toolkits in practice are configured with the implicit notion that customers know exactly what they want, feel comfortable with choosing from a limited number of options for product characteristics and use toolkit just as a communication interface. From a scientific prospective, most research focuses on investigating how a particular feature of a toolkit, such as a starting solution (Hildebrand, Häubl and Herrmann 2014), pre-configurations based on preference articulation (Boller, Schlager, Franke and Herrmann 2016), or peer-to-peer support by communities (Franke, Keinz and Schreier 2008; Jeppesen 2005) affects user satisfaction and preference fit. Our study differs from other studies significantly because we do not look at a particular toolkit feature, but propose a rather radical innovation. In the research project at hand, we investigate a radically different approach to toolkits structures a toolkit where customers have a possibility to freely design their product. Though such toolkits are not common in practice yet, this setup will be enabled through innovative technologies offered, for example, by MIT (Sketching Magic Paper) 135

149 or Google (Autodraw). More concrete, our research question will investigate whether this radically new toolkit architecture leads to a higher user satisfaction, preference fit, and willingness-to-pay. This is based on the assumption that creative talent is normally distributed and, therefore, all people have an inherent creativity talent to a certain degree (Osborn 1963). In such a way, the new toolkits only draws on the creativity that people naturally possess. The relevance of this research project is obvious as numerous scholars and practitioners have called for more research on the value created in self-design processes and the determining toolkit features (Franke and Hader 2014; Franke, Keinz and Schreier 2008; von Hippel and Katz 2002). By the 1st July 2017 we expect to finalize the literature review regarding problem-solving to determine the solid theoretical base and evaluate the technical programming options. Moreover, we envisage to complete a script for the first line of studies, which includes the experimental set-up, specification of the constructs that need to be measured, and the operationalization of variables. 136

150 Application of Structured Methods for Understanding Problems by User Innovators Poster Jennifer Otitigbe (Massachusetts College of Art and Design) Structured methods are commonly used in engineering, manufacturing, operations and healthcare as a way to guide individuals and teams through a process of understanding issues or problems and developing solutions and plans to address those issues (Ulrich and Eppinger, 1995). Structured methods vary in their goals, general familiarity, complexity and flexibility. One of the more commonly known structured methods for understanding problems is called a cause and effect diagram or an Ishikawa Diagram (Mehta, 2014). A cause and effect diagram is a simple way to visually brainstorm on causes related to an issue, observation, problem or event. While many people are familiar with cause and effect diagrams in the industries mentioned earlier, these diagrams are not very common beyond these functional areas unless the area or team has had an exposure to quality management principles. Generally, for this type of analysis, causes and sub-causes are arranged into standard categories (e.g. people, process, equipment etc.) relative to the effect, and the relationships between causes are represented as lines with arrows. This tool can help individuals and teams to understand why and to systematically generate a deep and broad understanding of the issue as a starting point for designing effective solutions (Katila and Ahuja, 2002). Cause and effect diagrams are just one of many tools used in problem solving and root cause analysis. One motivation for this study was an awareness that while individuals recognize that solving the right problem is more desired than solving any problem, few individuals or teams spend the time to ensure that they fully understand a problem before working on solutions (Spradlin, 2012). User innovators need to explore ways to structure their search for information to better understand problems. Would-be user innovators may be limited in their ability to understand complex problems, where the "stickiness" or difficulty of transferring or obtaining useful information may become a limiting factor due to their experience, background 137

151 or role within an organization (von Hippel, 2002). Innovation researchers acknowledge that Problem solving underlies innovation and that there exists a need to use the techniques of problem solving to design better solutions. (von Hippel 2016). In reference to the value of applying structured methods, researchers have observed that, for example, in the field of communications, structured communication methods are known to increase information processing efficiency by up to 40% (Abrahams, 2014). It is within this context that the use of structured methods by user innovators becomes a compelling research subject. This study examines three case studies where user innovators applied the cause and effect structured method to develop insights about specific issues as a part of a formal problem understanding, and eventual problem solving process. The research for this study involved conducting a literature scan for problem understanding and problem solving experiences, filtering those experiences down to usage examples of cause and effect analysis, and applying additional criteria to extract case studies where the structured method was used by user innovators, or individuals impacted by the problem under review. While many different structured methods exist, the goal of this research study is not to evaluate what methods work best within the context of problem understanding or problem solving, but rather, the goal is to uncover any insights or lessons learned regarding user innovators who applied structured methods themselves. Narrowing down to one specific structured method provides a basis for comparing and contrasting the case studies. In fact, there have been numerous articles that discuss positive and negative aspects of various problem understanding approaches (Anderson and Janson, 1979; Ramakrishna et al., 1986). Cause and effect diagrams are recognized as a problem understanding approach that is relatively easy to learn and recall (Anderson and Janson, 1979). This fact was one of the key reasons this study is anchored around cause and effect analysis. Beyond the usages of the diagram by user innovators, other criteria for selecting the case studies included; achieving a mix of disciplines of the user innovators, understanding motivations for using the method, and knowledge of outcomes. One finding that emerged from this preliminary research is that when user innovators apply a structured method, they are likely to modify it to better fit their goals and context. And perhaps the most important finding is that structured methods do give user innovators a framework to search for new information to improve their understanding of the 138

152 problem better than the case where they did not use the method. EXPECTED PROGRESS BY 07/2017: By July 2017 this research will be completed for the target set of three case studies in order to present a final set of outcomes and findings from the study. 139

153 User Innovation and Diffusion 140

154 Diffusion of User Innovation Within Online Communities from The Social Network Perspective Paper Wonho Lee (KAIST Graduate school of Business), Youngbae Kim (KAIST Graduate school of Business) Market failure of user innovation has been identified in several prior studies where the diffusion rate of user developed innovation was severely low (de Jong, von Hippel, Gault, Kuusisto, & Raasch, 2015; von Hippel, DeMonaco & de Jong, 2016). Lack of diffusion within an online community can jeopardize the potential innovative performance of the community where accumulative innovation is at the core of online user innovation activities (Watts & Dodds, 2007). This study focuses on user innovation diffusion (in terms of diffusion speed and diffusion rate) in the context of online community by applying the social structural influence of user innovators. Content analysis for 594,236 interactions between 3,489 community members was conducted. According to the findings, user innovations developed by an individual with higher in-degree centrality were found to have a significant relationship with diffusion speed. The strength of diffusion speed rapidly diminished beyond 3 days of introduction. User innovations developed by a community member with higher betweenness centrality were found to have a positive significant relationship with diffusion rate. The results of this study indicate that innovation diffusion is affected differently depending on the network positions of the user innovators in terms of diffusion speed and diffusion rate. 141

155 Why New Products Appear in Unexpected Places and What We Can Learn from It to Spur Product Development Paper Konstantin Fursov (National Research University Higher School of Economics), Jonathan Linton (National Research University Higher School of Economics) The need for appropriate infrastructure, human capital, and motivation for product innovation is considered. In doing so it is demonstrated that no one particular economic or political system is necessarily preferable to encourage new product development. Consideration of environments that have proven successful for innovation suggest that it can be independent of the presence of firms and/or inventors driven by the opportunity to capture the surplus economic benefit. While profit is a strong motivator for product development, it remains a proxy for the fulfillment of needs. New products may result as they lead to the fulfillment of one or more of these needs in the absence (or presence) of additional economic motivations. Four cases are provided to illustrate situations in which innovation has been obtained in the absence of a market system and a clear path for the inventor to capture the benefit. The components that support the development of an innovation rich environment are catalogued and four exemplars are used to illustrate this and provide saturation of the constructs and their inter-relation. It was demonstrated that to obtain new product development and innovation in the absence of profit motivation, the link between Need, Motivation, Enabling Technologies and Infrastructure must exist. In doing so, firms can determine non-financial models to obtain desired access to enabling technologies or infrastructure through the satisfaction of non-economic needs of partners. Governments can identify opportunities through citizen involvement to create social benefits. Prioritization of infrastructure development such as wireless access and enabling technology such as programming skills can be made with the development of innovation and new product development in mind. 142

156 The Field of External Search and the Search for External Knowledge Paper Sara Heuschneider, Daniel Ehls (Hamburg University of Technology), Cornelius Herstatt The search for new knowledge represents a core aspect to increase organizational performance and achieve a competitive advantage. By engaging in external search, organizations strive to utilize distributed knowledge sources to enrich their knowledge base. While research attention on external search has been strong over the past years, it appears that scholars prefer to concentrate on rather isolated bodies of literature and miss connections among them. Against this background, we aim to analyze the current research on external search and to identify the intellectual pillars of the field as well as to provide a landscape of different research streams. By conducting an extensive literature co-citation and content analysis of relevant publications, we are able to reveal a fragmentation of the research field and disjoint schools of thought. We uncover the intellectual pillars, their linkages to each other, and the theoretical foundation of the research field on external search. Our findings help scholars to navigate on the complex topic landscape of external search and provide implications and future research directions to further drive the development of the field. 143

157 Decision Makers Underestimation of User Innovation Paper Philip Bradonjic (Vienna University of Economics & Business), Nikolaus Franke (Vienna University of Economics & Business), Christian Luethje (Hamburg University of Technology) In this study, we investigate if and to what extent important decision makers underestimate the value of user innovations. Knowledge on this phenomenon is important, as it points to a perceptual bias with potentially critical consequences on the allocation of resources, on innovation dynamics, and on social welfare. We capture underestimations with a newly developed and tested scale. Our core finding is that decision makers actually underestimate user innovation to a large degree. The core reason is knowledge on innovation management: The more decision-makers know, the more they underestimate user innovation. In a second study (ongoing), we explore the root causes for this finding. It appears that information diffusion agents (such as textbooks and media) report about the sources of innovation in a particularly biased way: They hardly mention users as possible innovators. Thus, more knowledge on innovation is knowledge that creates a bias. These findings have implications for business and public administration. 144

158 User Innovation A Systematic Literature Review from 1970 to 2013 Paper Sandra Jennina Sanchez (Universidad EAN), Alazne Mujika-Alberdi (Universidad de Deusto) From the late 1970s onwards, a considerably body of literature related to user innovation has been published. This study provides a rigorous systematic literature review of the field. The sample was 146 articles. The results show the evolution of several research streams, the most prominent authors, academic journals and countries. 145

159 Innovation Diffusion via Remixing Digital Objects: The Role of Wordof-Mouth Commentary in Online User Communities Paper Gregory Fisher (Miami University), Michael Stanko (North Carolina State University) Remixing digital objects within online user communities is a contemporary form of innovation diffusion that allows creative works to propagate while taking divergent directions. Remixing occurs frequently in user communities devoted to literature, photography and even physical objects. Managers are keen to engage with customers through remixing to bolster new product development efforts and to deepen their relationship with customers. Within online communities, users comments regarding specific digital objects have the potential to impact how the community engages with such objects through remixing. Extant wordof-mouth research has generally found that negative commentary dampens consumer enthusiasm about products and brands. However, our research leverages two large datasets collected from online user communities devoted to innovation and multimedia to empirically show that negative commentary promotes community remixing (to a point). Negative commentary provokes questions that lead to the generation of user solutions through remixing. These findings extend innovation diffusion theory to better understand remixing. 146

160 Opportunity costs of Household Sector Innovation A study in the Emirates Paper Daan Rademaker (Utrecht University School of Economics), Jeroen P.J. de Jong (Utrecht University School of Economics) The main objective of this paper is to empirically test whether or not opportunity costs faced by users are a significant predictor of likelihood of a user engaging in user innovation process. An opportunity cost framework is often used in economic decision making models. In these models individuals optimize their allocation of time between several alternatives based on their preference in substituting between these alternatives. Entrepreneurship studies show that opportunity costs matter for the decision to start a business: at higher wages, given certain levels of ability, individuals were less inclined to venture. In this study we build on this line of thinking by proposing that, when individuals face a particular need, given their abilities to solve this need they will evaluate whether the time investment (effort) concerned with innovation outweighs the benefits of time dedicated to alternatives. In the study we focus on wages as a proxy for these opportunity costs. We use survey data on household sector innovation from the United Arab Emirates to estimate the effect of wages on likelihood to innovate. The dataset includes 3000 individuals who have taken a phone survey, of which a fraction (around 5%) has engaged in household sector innovation in the past three years. To fix the effect of innovation ability, we will use a matching procedure where based on gender, technical skill and education. We look at pairs of individuals whom are similar on these demographic variables, but diverse in terms of their innovation outcome (innovator or not). We will then see if on average the difference between these pairs is significant. At the OUI Workshop, first findings will be briefly presented. 147

161 User Innovation and Diffusion in Online Firm-hosted User Communities The Case of China Poster Yu-Shan Su (National Taiwan Normal University) 1. How user innovations become commercial products? In the studies of the source of innovation, successfully commercialized products are often based on user-developed innovations (Baldwin et al., 2006; von Hippel et al., 2011; von Hippel et al., 2012). Many of user innovations can generate general value. In developing commercialized products, what one consumer requires may fit what another wants better. When user innovations are valuable to others, the diffusion of user innovations by individuals can enhance social welfare (Gambardella, 2017). Hence, it s potential to increase innovative users to understand the general value of their developed innovations and increase their incentives to engage in commercial diffusion. Established firms have an incentive to search for adopt user designs and user innovations of potential commercial value for new product concepts and prototypes (Gambardella, 2017). Producers of consumer products systematically seek to consumer-developed innovations can learn more effectively and use consumer-developed innovations as free inputs to their own product development efforts (von Hippel et al., 2011; von Hippel et al., 2012). Companies may also adopt many innovation management and practices and assistances to make related information of user innovations available (Gambardella, 2017), such as finding and benefiting from lead user innovators (Lilien et al., 2002; von Hippel, 1986), offering innovation toolkits to consumer (Thomke and von Hippel, 2002), holding innovation contests of consumer activities (Jeppesen and Lakhani, 2010), and building on-line communities for users (Dahlander and Frederiksen, 2012; Franke and Shah, 2003; Jeppesen and Frederiksen, 2006). 2. A paradigm shift: from producer innovation to user and open collaborative innovation Uses collaborative innovations enhance both commercial diffusion and peer-to-peer diffusion in user innovations (Baldwin and von Hippel, 2011; de Jong et al., 2015). Baldwin and von Hippel (2011) proposed Modeling a paradigm shift: from producer 148

162 innovation to user and open collaborative innovation. Baldwin and von Hippel (2011) argued that user innovators collaborating and receiving helps from other users reduce innovation costs and diffuse user innovations partially and fully in the process of users collaborative innovation process. In this case of users collaborative innovations, collaborations generate and facilitate the diffusion of user innovations for both user innovators and free rider adopters at the same time. Baldwin and von Hippel (2006) argued one or more users join into communities and motivate by the increased efficiency of collective innovation, as user manufactures emerge and enter. In Ogawa and Pongtanalert s (2013) research, they found the innovation adopt rate by peers was 48.5% as innovative users belonged to communities with a shared interest in their developed innovation; otherwise, the innovation adopt rate by peers was 13.3% as innovative users did not belong to communities. On the one hand, community users tend to collaborate and assist one another, including sharing information on innovations that they have developed among end-user (Franke and Shah, 2003). On the other hand, users freely reveal innovations to a firm s product platform in online firm-hosted user communities, which can contribute and put the firm in a favorable position (Jeppesen and Frederiksen, 2006). 3. The Case of China Xiaomi and Haier In online firm-hosted user communities, users often interact and help to each other solve problems in communities. In China, leading firms of consumer products create user innovations and diffusion by online firmhosted user communities. I use Beijing Xiaomi Technology Company and Haier Corporation as two cases of online firm-hosted user communities in China. In China, Beijing Xiaomi Technology Company and Haier Corporation employ online communities of users to facilitate and strengthen their innovation process and outcome. There are many scholars contribute their knowledge to the research of user innovation in online firm-hosted user communities. Dahlander and Frederiksen (2012) showed how the online communities and the type of innovative users behavior enable the effects of innovation. Von Hippel (1986) suggested that organizations focus on selected contributors, while an organization can seek suggestions from individual contributors in its network. Innovative users are likely to be lead users, who help to discover commercially attractive user innovations (Von Hippel, 1986; Jeppesen and Frederiksen, 2006). There are two reasons for two Chinese leading firms of consumer products adopting online firm-hosted user communities. The first, distant search gives 149

163 organization access to knowledge that may not reside within their organizational boundaries (Afuah& Tucci, 2012; Jeppesen and Lakhani, 2010). The second, crowding narrows the attention of the organization; that is the organizations are more likely to pay attention to the suggestions that are familiar, not distant, though organizations efforts reach out to external contributors and access suggestions that capture distant knowledge (Piezunka and Dahlander, 2015). Two leading Chinese cases of Beijing Xiaomi Technology Company and Haier Corporation are shown. 150

164 Measuring Innovation Performance with Creativity between Free Closed and Open Innovators Poster André Witzel (RWTH Aachen University) In the last decades, new innovation research streams divergent from the classical theory of the sole producer innovator (i.e. closed innovation; Schumpter 1934) were born from the socioeconomic changes like the emergence of the internet. Especially two streams: User Innovation (von Hippel 1988, 2006) with the focus on innovations by individual user or individual user firms and Open Innovation (Chesbrough 2003) with the focus on producer opening up their borders and that the locus of innovation development and production do not necessarily must be the same. With his latest description of the Free Innovation Paradigm von Hippel (2016) brings in a new stream of innovation research. Fulfilling the already postulated paradigm shift (Baldwin & von Hippel 2011) away from the producer as the sole innovator towards the customer into his own format. Within the paradigm, he identifies the household sector as source of innovation (von Hippel, de Jong, & Flowers 2012; von Hippel, Ogawa, & de Jong 2011; de Jong, von Hippel, Gault, Kuusisto, & Raasch 2015; de Jong 2013; Kim 2015), which by definition innovates for non-monetary reasons. In this study, I like to analyze whether and to what extent the creativity of innovation by free innovators is higher or lower than that of producer innovators (i.e. closed and open innovation)(see Figure 1). The choice for creativity as a measure of innovation performance becomes clearer with the product-oriented definition of creativity: something novel that is useful (Stein 1974; Mumford 2003). On the one hand, with the official definition of innovation in mind (Oslo Manual 2005, paragraph 146), both, a creative product and an innovation, need to have a certain degree of Novelty. On the other hand, a creative product needs to be useful. Usefulness can be defined as [ ] the degree to which a product enables a user to achieve his or her goals, and is an assessment of the user s willingness to use the product at all (Rubin & Chisnell 2008). From this, it can be deduced that usefulness may also be an indicator for commercial success. As the self- 151

165 reward of personal use (von Hippel 2006) and the later free adoption imply that there is a general interest, usefulness should be significantly higher within the group of free innovators. In particular, studies on lead users show that innovations from self-rewarded users can lead to commercial success (von Hippel 1986; Urban & von Hippel 1988; Franke, von Hippel, & Schreier 2006; Hienerth, von Hippel, & Jensen 2014). In order to measure and compare the creativity of different actors, comparable products and services are required. For that purpose, I identified a community of consumer goods (i.e. board games). Within this community, both free innovator and producer innovator (i.e. closed and open innovation) can be observed, developing and making new products available for diffusion. 152

166 User Innovation and Psychology 153

167 The Company or the Crowd? Comparing Consumers Reactions to Peer-Provided and Firm-Provided Customer Support Paper Lan Jiang (University of Oregon), Matthew O'Hern (University of New Hampshire), Sara Hanson (University of Richmond) This research demonstrates that when a consumer encounters a product-related problem, peers have a unique advantage in offering social support to their fellow consumers and this social benefit can be a dominant driver of consumer satisfaction, following an unsuccessful problem-solving attempt. Using real world data from an online support community, a pilot study finds that when a problem goes unsolved, satisfaction is greater when consumers receive peerprovided vs. firm-provided support. Study 1 replicates this finding in a controlled experiment. Study 2 identifies social support as the mechanism that underlies this effect and investigates whether firm employees can take steps to appear more customer-like and thereby replicate the advantage of peer-provided support. Finally, Study 3 reveals that mimicry is another way for firms to enhance social support. Specifically, when a firm employee imitates the language or paralanguage (e.g., avatars, emojis) that the consumer used to describe the focal problem, feelings of social support increase, resulting in greater satisfaction. 154

168 Uncovering the Value Basis of User Innovation Poster Helle Alsted Søndergaard (Aarhus University), John Thøgersen (Aarhus University) Users as innovators has gained widespread interest since von Hippel s seminal work in the 1970s (Hippel, 1976). Instead of merely being viewed a homogeneous and passive group, users became understood as a heterogeneous and active group, that is the source of many novel concepts and ideas (von Hippel, 1986). Since then, a whole stream of research on users, especially lead users has emerged (Hienerth & Lettl, 2016). After establishing the importance and prevalence of user innovation, recent research has estimated the proportion of ordinary consumers that create innovations (Kim, 2015; von Hippel, Jong, & Flowers, 2012; von Hippel, Ogawa, de Jong, & Jong, 2011). This research has found, among other things, that male consumers that are highly educated and have technical training are more likely to be user innovators (von Hippel et al., 2011). However, apart from such demographic characteristics, and the fact that users in general innovate for non-economic reasons, such as personal need, fun, learning or helping others (von Hippel, 2016; de Jong 2016), we still know little about consumer innovators and how innovative behavior at this level can be predicted and supported. The aim of this study, which is built on existing user innovation research, is to investigate whether the value priorities of consumer innovators differ from others. We wish to more precisely identify and profile consumer innovators by exploring their value priorities compared to noninnovating consumers. For this purpose, we use a survey of consumer innovation across ten European countries. For this study, we build on user innovation theory and theory of human values. Users were identified by von Hippel as a source of innovation already in the 1970s and by now the characteristics of especially lead users are well tested and described (Hienerth & Lettl, 2016). Research on lead users find that they have a higher level of leading edge status (Morrison, Roberts, & Midgley, 2004), create more attractive innovations (Franke, von Hippel, & Schreier, 2006), and get more novel and original ideas that lead to better performance (Lilien, Morrison, Searls, Sonnack, & von Hippel, 2002). With 155

169 the aim of documenting the importance of user innovation for society, recent studies have established the prevalence of normal consumers innovating by identifying the proportion of households that engage in user innovation (de Jong, 2016; Kim, 2015; von Hippel et al., 2012, 2011). Recently, other characteristics of user innovators than demographics have begun to be investigated, revealing that certain personality traits impact the success of consumer innovations in the different stages of the innovation process (Stock, Von Hippel, & Gillert, 2016). From the big five personality traits, openness to experience has been found to be positively related to the ideation phase, while introversion is positively related to prototyping and conscientiousness to the diffusion phase (Stock et al., 2016). However, a broader range of antecedents that may help to predict and support consumer innovation is yet to be established. Regarding human values, this study is based on Schwartz value theory, which defines values as desirable goals, varying in importance, that serve as guiding principles in people s lives (Schwartz, 1992). Based on theorizing and value surveys in a large number of countries from all inhabited continents, Schwartz (1992, 1994) identified 10 universal value types, or motivation domains. As a result of their mutual consistency and conflicts, these 10 universal human values form a circumplex, which can be mapped onto a twodimensional space with the two main dimensions representing selftranscendence versus self-enhancement and openness-to-change versus conservation. A survey study was carried out in 10 European countries: Germany, Netherlands, Denmark, Finland, Italy, Spain, Poland, Hungary, France, UK (N 1000 in each country), covering the five regions north, south, east, west and central Europe. The survey contained questions about human values, consumer innovation activities and a range of relevant background characteristics, as well as questions not pertinent to the present study. The questionnaire was developed in English and translated to the nine other national languages. In order to control the translations, they were backtranslated into English by the organizations doing the translation (but a different person). The authors controlled the back-translations, comparing them with the original English version and settled all uncertainties and ambiguities with the translators. Before implementing the surveys, the final, online questionnaire was further controlled by a knowledgeable, native speaker of that language, which led to a few, minor corrections. The survey data were collected 156

170 by a professional market research company (YouGov), who sampled respondents from their own and partners panels in the different countries, administered the data collection as CAWI-interviews and organized and presented the data in SPSS files. The samples from each country are representative for the age group years old with regard to gender, age and geography. The open responses to the consumer innovation questions were thoroughly analyzed with the assistance of native speakers to check if reported consumer innovations are really user innovations, following the procedure and criteria developed by de Jong and von Hippel (de Jong, 2016; von Hippel, de Jong, & Flowers, 2012). Based on these responses, participants were classified as user innovators or not. This categorization was regressed (logistic regression) on a range of individual characteristics in two blocks using a stepwise procedure: Block 1: country of origin, gender, age, income. Block 2: Schwartz s four main value dimensions, corrected for general response tendency. Results We find that the likelihood of being user-innovator: a) increases with openness-to-change and selftranscendence values and b) decreases with conservation and self-enhancement values The likelihood also depends on country of residences and gender, but not on age or income. Result tables and statistics will be shown in the final presentation. 157

171 Lead Userness and Innovative Work Behavior in Application Development: A Dual Path Model Poster Mario Schaarschmidt (University of Koblenz-Landau), Dirk Homscheid (University of Koblenz-Landau), Björn Höber (University of Koblenz-Landau), Matthias Bertram (Provadis School, Frankfurt) The innovation literature has long maintained that users of products and services develop specific knowledge sets concerning the actual use. Thus, users combine need and solution knowledge. For innovating firms, such knowledge is of crucial importance as it complements own search for solutions in internal R&D departments. Not all users are prone to innovate, that is, not all users are aware of their need-solution pairs or are simply not willing to either share the knowledge with a company or even to start own innovation activities. Those who do are usually considered to be lead users. In this research, we aim to test how two perspectives, namely the efficiency-oriented perspective, and the social-political perspective, help explaining why lead userness is associated with innovative work behavior. As a context for this study we chose application development, as it is a highly innovative industry, characterized by a large volume of development activity by volunteers as well as paid developers. Thus, in addition to the main purpose, it is suitable to study differences for paid (embedded) lead users and volunteers. 158

172 The Interplay of E-Lancers Character Traits and Digital Signals Performance Implications in Online Labor Markets Poster Ruth M. Stock (Technische Universität Darmstadt), Christian Holthaus (Technische Universität Darmstadt) Recently, there has been a growing body of research on online labor markets - online platforms for individuals and organizations to quickly identify shortterm workers [e-lancers] who have skills required for particular, often one-time, tasks. (Farrell et al. 2016, p. 1). Examples for online labor markets are freelancer.com and upwork.com. In these online platforms, e-lancers are required to send diverse signals to the market that reduce information asymmetry and attract new customers. We present a first of type study to explore how e-lancers individual character traits may influence this signaling behavior and their performance in online labor markets. Accordingly, we distinguish between different types of traits, such as extraversion and openness to experience as well as different types of signals those that are easy to fake (conventional signals) and those that are hard to fake (handicap signals). Various types of outcomes are assessed such as customer satisfaction, profit and volume. This study is the first to distinguish both types of signals in online labor markets and assess their antecedents and outcomes that play a central role for the daily business of e-lancers and their clients in online labor markets. We use a distinct multisource dataset consisting of 1120 e-lancers from a major online labor market platform to cater the needs of our study design. We combine survey data, independent coders assessment and objective performance data in online labor markets. Thus, we obtain self-report measures of character traits, objective measures of performance and various handicap signals as well as measures of conventional signals as perceived by third raters. 159

173 Does Job-related Innovation Benefit User Innovations? An Investigation of Spillover Effects Poster Ruth M. Stock (Technische Universität Darmstadt), Carmen S. Lukoschek (Technische Universität Darmstadt) An Investigation of Spillover Effects Professor Ruth M. Stock, Ph.D. Technische Universität Darmstadt Germany & Carmen S. Lukoschek, Ph.D. Candidate Technische Universität Darmstadt Germany In contrast to the traditional producer-innovation-paradigm, empirical studies show that users are a major source of innovations (e.g., Gambardella et al. 2016). Research has provided considerable evidence about the economic and social benefits gained from these user innovations (e.g., de Jong et al. 2016; Gambardella et al. 2016; von Hippel 2017; Stock et al. 2015). Yet, surprisingly little is known about the reverse side of the coin - that is, whether and how user innovators benefit from their social and working environment. In an attempt to bridge this gap, the present study focuses on user innovators work environment. Specifically, we provide insights into the relationship between innovativeness as a job requirement and users perceived learning benefits for their user innovation. Innovativeness as a job requirement is the extent to which the relevance of innovative behavior is explicitly specified in a person s job requirements (Yuan & Woodman 2010). Perceived learning benefits for the user innovation is the extent to which user innovators derive an advantage for their user innovation from learnings in their jobs (self-developed). In addition, the study sheds light on how these learning benefits translate into increased quality of the user innovation. Accordingly, we seek to answer the following research questions: (1) Does innovativeness as a job requirement affect user innovators perceived learning benefits for the user innovation? (2) Do perceived learning benefits affect the quality of the user innovation? (3) Does job experience moderate the relationship between innovativeness as a job requirement and perceived learning benefits? Relying on learning theory and social cognitive theory (e.g., Bandura 1986; Kolb 1984), our study builds upon the premise that acquisition and development of new 160

174 knowledge depends on individuals interacting with their environment and actively observing, replicating, and developing rewarded behaviors. Accordingly, we propose that user innovators, employed in jobs that demand highly innovative behavior, experience spillover effects through which job-related insights affect the quality of their user innovation. Thus: H1: Innovativeness as a job requirement positively affects user innovators perceived learning benefits for the user innovation. H2: Perceived learning benefits for the user innovation positively affect the quality of the user innovation. H3: Job experience strengthens the relationship between innovativeness as a job requirement and perceived learning benefits. To test our hypotheses, we collected data from 120 user innovators, including descriptions of their user innovation and information on the independent variables, as well as company-related and demographic control variables. To assess the quality of the user innovations, three independent raters evaluated the innovations with regard to their meaningfulness (adapted from Im & Workman 2004), novelty (adapted from Miron-Spektor & Beenen 2015), and technical goodness (inspired by Amabile et al. 1996). SEM analyses in MPLUS 7.2 (Múthen & Múthen 2015) indicate satisfactory fit for global measures (RMSEA = 0.047; SRMR = 0.056; CFI = 0.985; TLI = 0.979). In support of hypothesis H1, innovativeness as a job requirement positively affects perceived learning benefits for the user innovation (ß = 0.33; p < 0.001). In support of hypothesis H2, perceived learning benefits positively affect the quality of the user innovation (ß = 0.23, p < 0.05 for meaningfulness; ß = 0.23, p < 0.05 for novelty; ß = 0.29, p < for technical goodness). Moreover, multiple regression analyses provide first support for hypothesis H3. We can find a moderating effect of job experience on the relationship between innovativeness as a job requirement and perceived learning benefits (baseline model: ß = 0.40; p < 0.001; R 2 = 0.19; p < 0.001; moderation: ß = 0.22; p < 0.01; R 2 = 0.23; p < 0.001; DELTA R2 = 0.05, p < 0.01). In our final paper, the moderator analyses will be consolidated into a coherent SEM model. Our insights provide important implications for research and policy makers. First, our results indicate that user innovators can benefit from learning-induced spillover effects of their work environment. To further increase the number of user innovations, policy makers may hence consider incentivizing companies that foster employees engagement in innovative behaviors. Via collaborations with these companies, policy makers could provide information and workspaces that support the 161

175 development of user innovations among employees in a more targeted fashion. Moreover, the results show that increased job experience positively affects the quality of the innovation through perceived learning benefits. This may yield a convincing argument to policy makers encouraging companies to hire older, more experienced employees as they may be more capable of transferring knowledge from previous innovation projects to new work-related projects that is analogous to the spillover effects observed from work-related to user innovations. Third, as a first-of-kind study, we open the field for research into benefits user innovators may gain from their immediate and work-related environment for the development of their user innovation. Hence, we invite research to complement the current view on benefits of user innovations to include benefits for user innovations that may encourage more individuals to join the user innovator community. 162

176 From User-Innovator to User-Entrepreneur: Designing and Delivering a University Course to Foster the Co-Creation of Business Models Poster Albrecht Karlusch (Vienna University of Economics & Business), Kathrin Reinsberger (Vienna University of Economics & Business), Wolfgang Sachsenhofer (Vienna University of Economics & Business) History has shown that users are often better in inventing new and innovative products compared to large corporations (von Hippel, 1986). User innovators typically focus on the personal use of an invented product and do not aim at commercializing this innovation (Shah & Tripsas, 2012) by means of a proper business model (Chesborough and Rosenbloom, 2002). At the same time, startups are very successful in finding explorative business models (Chesbrough, 2010). The development of such an explorative business model is positively affected by interdisciplinary skills, lateral thinking and incorporated intelligence (Cardon & Stevens, 2004; Ruef, Aldrich, & Carter, 2003). Fostering the creation of an interdisciplinary start-up on the top of a user innovation might therefor be an effective way to better realize users existing innovative potential. Universities around the world have started to design curriculums and courses in entrepreneurship to foster innovation and the creation of explorative business models in terms of quality and quantity (Martin et al., 2013). The majority of published cases are descriptive and mainly focus on business- or technical-aspects and omit to consider other relevant abilities, such as creativity (Lozano, Ceulemans, & Scarff Seatter, 2015). Only a limited number of the presented cases are based on teaching and learning-theories.however, integrating the skills of artists can stimulate the establishment of a more critical, innovative and reflective culture that frequently questions its own routines, assumptions and guiding principles (Lans et al., 2014). This paper presents a case (Yin, 2014) based on the process of designing and delivering a course for the co-creation of explorative business models based on users own innovative ideas at a Business school together with a University of Applied Arts. The process is based on four 163

177 key elements: (1) the course objectives; (2) the learning outcomes referring to cognitive and affective learnings as well as to the co-created business model; (3) the course structure, representing the didactic implementation; and (4) the course assessment, conceptualized with triangulated assessment methods. Method The developed course relies on social learning theory (Bandura, 1971; Ferguson & Buckingham Shum, 2012), the concepts of user innovation (Ekeh, 1974; Lakhani & von Hippel, 2003; von Hippel & Krogh, 2003, von Hippel, 2005) and a constructivist position (Colliver, 2002) to address both, the creation of effective interdisciplinary student-teams (Beckman, 2006; Eisenhardt & Schoonhoven, 1990) and the complexities of co-creating an explorative business model based on a user innovation. Course design & delivery The course objective is to educate business- and art-students as interdisciplinary agents, able to understand the different perspectives, combine the various approaches from a holistic point of view and co-create together an explorative business model on the top of the user innovation. The learning outcomes are segmented into a cognitive-, affective- and skill-based dimension (Kraiger, Ford, & Salas, 1993). The cognitive dimension includes declarative expert knowledge about new technologies as well as methodological knowledge about user innovation (von Hippel, 2005), business-modelling, the Lean Start-up method (Blank, 2013) and the Stanford Design Thinking approach (Brown, 2008). The affective dimension is linked to activities to reduce mental and emotional barriers between students with very different backgrounds. The skill-based dimension addresses students abilities to overcome affective barriers between each other, create effective teams and use their creative potential to co-create explorative business models on the top of user innovations. One of the major pedagogical challenges is to design a didactic course structure in which the art- and business-students are able to quickly collaborate despite their huge differences in culture, language and habit. A number of course elements, such as exercises to explicate the specific norms and values, controlled debates and social group challenges based on the person-centred approach (Fielding, 2006) were included to quickly reduce emotional barriers. Thereafter the course design consists of an iterative alteration of a dreaming- and an evaluation-phase, similar to the trial-and-error cycle of user innovations (von Hippel, 2005). The abilities of art students are particularly valuable in the creation phase while the know-how and analytical abilities of business students are more relevant in the evaluation phase. These 164

178 two very different emphases ensure that both groups, artists and businessstudents can increase their reputation and add value to the project (Lakhani & von Hippel, 2003; von Hippel & Krogh, 2003, Wasko & Faraj, 2005). The course assessment includes the evaluation of students learning progress, the feedback from students and an external valuation of the project outcomes by experts. Results Preliminary results of this study strongly support the assumption, that linking business with art students in a supportive environment has a very high potential to foster innovation and the co-creation of explorative business models based on user innovations. By July the 1st we expect to have first final results on the effect of such a course on the transition from the user innovator to the user-entrepreneur. 165

179 User Innovation and Healthcare 166

180 A Next Generation for Public Health Intervention Models: The Public as Innovators Paper Christiana von Hippel (Harvard T.H. Chan School of Public Health) In the public health field, the design of public health interventions has long been considered to be the province of public health experts. In this paper, I explore an important complementary approach: the design and prototyping of public health interventions by consumers themselves. In the approach I propose, consumer prototype interventions are systematically identified and evaluated by public health experts, who then design improvements and diffusion strategies as appropriate. The context and support for my proposal builds upon research by many innovation scholars that has documented a strong pattern of consumer innovation in the development of new consumer products. As I will explain, product producers have learned to profit from these grass roots designs and field experiments. Instead of just searching for unfilled consumer needs, leadingedge firms are increasingly searching for consumer-developed solutions to their needs as the basis for profitable new commercial products. I explore the applicability of this consumer innovation pattern to in the public health field. I first document its potential effectiveness via published examples of its use in public health in the form of positive deviance studies. Next, I discuss the potential of a public health intervention development model that builds upon more efficient, lead user methods for identification of consumer innovation. Finally, I propose that research and experimentation by public health researchers on the promise of consumer-prototyped health interventions will enable us to develop new methods to more efficiently and effectively intervene on preventive health problems facing the communities we serve. 167

181 People with Disabilities as Product Innovators: A Pilot Study Paper Peter Conradie (Ghent University), Aron-Levi Herregodts (Ghent University), Lieven De Marez (Ghent University), Jelle Saldien (Ghent University) The purpose of this study is to explore the rate of product innovation among persons with disabilities, how they are diffused, and the general value of these developed solutions for persons without disabilities. A sample of participants (n = 178) completed a self-administered questionnaire, which included information about their disability, disability burden, general unmet product needs, the impact of the developed solution on their quality of life, and how solutions were diffused. We analysed solutions both for their novelty and for their general value. Close to 45 percent of respondents reported having developed a solution, with 9.55% solutions judged as novel. Additionally, 6.8% of respondents developed a solution that of general value to non-disabled users. Our results suggest that people with disabilities are actively involved in product development and that their solutions have a positive impact on their lives. Furthermore, many reported solutions also provide value for non-disabled persons, suggesting that persons with disabilities may be an important source of innovation. 168

182 If necessity is the mother of invention, are patients from developing countries keen innovators? Paper Pedro Oliveira (Católica Lisbon School of Business and Economics), Helena Canhão (Nova Medical School), Salomé Azevedo (Católica Lisbon School of Business and Economics), João Silva (Católica Lisbon School of Business and Economics) If necessity is the mother of invention, it makes sense to search for interesting and innovative solutions in contexts and geographies of high need. Healthcare in emerging economies combines necessity in two different ways. On one side, the type of need that is imposed on a patient by a disease or health conditions. On the other side, the economic context of the geography. This combination should, in theory, drive patients and caregivers to have characteristically higher-than- average-needs and consequently make then keen innovators. At lease in theory, as obviously users in those two contexts will also phase major obstacles both due to the restrictions imposed by the disease and the economic context. Analysing need as a determinant for the locus of innovation is especially relevant at a time when users are gaining more access to tools and technologies that allow them to solve problems in novel way. This study examines the extent to which patients and their caregivers, in developing countries, innovate, as well as the factors that enable these innovations. We also explore if these innovations are meaningful on a global stage. To study this issue, we conducted an empirical investigation of 44 health care innovations by patients and caregivers from developing economies and conducted 15 semi-structured interviews. We attempt to answer the following research questions: To what extent are patients and caregivers from developing economies able to innovate? Which factors enable these innovations? To what extent can user solutions, created in developing countries, be adopted in developed regions? 169

183 The Evolution of Stealth Innovation in Nursing: History, Drivers and Prototyping Genome from the MakerNurse Study Paper Anna Young (MakerNurse), Nikolas Albarran (Pop Up Labs), David Marshall (University of Texas Medical Branch), Maureen DeMenna (South Shore Hospital), Max-Philipp Schrader (Max Planck Institute for Innovation & Competition), Jose Gomez- Marquez (Massachusetts Institute of Technology) Making in health as the intentional activity when individuals shape, form, assemble and transform objects with their own hands-on skills and nearby resources. It represents the integration of a skillset, a toolset and a mindset. Nursing has a rich tradition of making--one that has produced numerous devices and tools that created a direct path to improved patient outcomes. For this paper, we looked back at more than 100 years of nursing history to uncover this rich tradition of creating and spreading tangible, nurse-made devices and discover how nurse making has evolved over time--including its links to nursing research and clinical innovation. We examine the transformation of making in nurse practice from it's heyday in the early part of the 20th century, a resurgence in the midcentury, and its transition to a state of stealth making. Nurses formed a network of innovators in history and in the present that are changing the way care is delivered, redefining the role of prototyping, and highlighting the importance of creativity over evidence as the source of innovation in practice. We introduce results of a US study analyzing the portfolio, barriers and drivers of nurse-made devices in hospitals and clinics today. 170

184 The Hurdles to Diffusion of a User Innovation in a Market Dominated by Experts The Case Of PEARS Poster Leid Zejnilovic (Nova SBE), Pedro Oliveira (Católica-Lisbon School of Business and Economics) Research shows that user entrepreneurs have founded 46 percent of the U.S. start-ups that have lasted 5-years or more. The researchers attribute the success of user entrepreneurs to the value user innovations they bring into business, a good understanding of the user needs, and a high level of uniqueness of their products and services that they manage to maintain and appropriate well. But, little is known of the hurdles the user entrepreneurs face on the path to commercial success which arise because of their non-expert origin. In this paper, we report on a case study in medical the sector, where the experts control the diffusion of an innovation and may be intentionally slowing it down. We first conduct a cost-benefit analysis to establish the relative value of the user innovation in comparison with market alternatives and explore the range of explanation for the diffusion failure. We find the user innovation in focus to be superior to its alternatives, in lower costs and 1-year outcomes for the patients. We establish the benchmark of the diffusion of a similar innovation, and identify the obstacles to diffusion of the user innovation. Our findings suggest that a combination of norms in the community of (practice) experts and institutional inertia are jointly slowing down the shift to a socially more desirable (user) innovation. We argue that this finding is generalizable beyond the case we analyze and discuss the options for the user entrepreneurs and policy makers in markets where experts control the diffusion process. Introduction Personalized external aortic root support (PEARS) is an award winning and, arguably, one of the most impressive user innovations in healthcare. Tal Golesworthy, a Marfan syndrome patient, came up with an idea for the aortic root support to fix his own heart, developed a computer-aided design based procedure for producing it, fundraised for the research team that developed a material suitable for the intended use, and managed the research 171

185 and development process. On the top of it, he established a firm to commercialize the product he developed to fix his heart ExoVasc Ltd. Over a decade of commercialization efforts later, the PEARS is repeatedly proving its success by flawlessly serving 78 Marfan syndrome patients, globally (as of May 2017). Still, PEARS is far from being widely adopted or even reaching a notable market penetration, considering the estimates of about ten thousand Marfan syndrome patients in the world, and about 110 individuals annually eligible for a root operation in the UK only (Treasure and Pepper 2015). The number of potential customers is higher, as the PEARS may be a surgical option for any genetically determined aneurysm of the aortic root with an estimate of around 2000 cardiac operations in Europe, annually, which could benefit from the implant. One obvious explanation for the low diffusion/adoption rate would be that the product is inferior to its competition. But, consider an alternative explanation; perhaps it is because a patient developed and commercialized a product/procedure that is aiming to replace widely accepted products/procedures developed by experts. If this alternative explanation is true, a more fundamental question arises - should we care if a product/service developed by a user is at low odds to succeed in a market where experts control the diffusion channels, and the quality of the innovation has little to do with its success in a market? Even a possibility of such a situation of a user innovation being slowed down or blocked by experts/producers controlling the diffusion channels, and not because it being inferior to the alternative requires developing an understanding and mechanisms to level the playing field. If experts decide to ignore a superior product/service at the cost of directly affected individuals and society, it seems right to question such a decision and find ways to provide a possibility of choice for patients. To explore this case, we first conduct a cost-benefit analysis of the PEARS and main competing procedures the dominant total aortic root replacement and aortic root sparing alternatives. We find that the PEARS dominates the alternatives; it is up to three times cheaper for a typical patient and 50-years product use basis, with lower risks during the intervention, and better 1-year outcomes. In other words, the inferiority of PEARS to competition does not hold as a reasonable explanation for the low diffusion rate. Tal Golesworthy, the innovator, explicitly stated the alternative explanation we explore in this paper. In an interview, he said, All I set out to do was to fix myself. I didn t set out to change the world, but I ve 172

186 found myself in a position where I m trying to change a part of the world that is very resistant to change. If I had any sense I d walk away, but it seems I don t have any. I can t abandon those patients who want to exercise their choice of treatment. We do not take this explanation at its face value. We use a case study analysis and further investigate the hurdles to a user entrepreneurship. We looked into the process of the product development, financing, the strategy for commercialization, the activities to sell and maintain the quality of service, and the expansion strategy. Key informants are Tal Golesworthy and a cardio surgeon is operating using the PEARS. Also, we analyzed public documents and interviews with patients and experts directly related to the PEARS. We identify two key generators of the obstacles to diffusion (experts and institutions) and the activities that reinforce the effect of these obstacles over time. We discuss the failures in policies that propel such unwanted outcomes, and the options for user entrepreneurs, policy makers, and the experts interested in a fair and socially responsible practice. This work adds to the theory of user entrepreneurship by looking into resistance to user innovation in markets where experts control the diffusion. It also contributes to the practitioners, by showcasing the issues likely to occur when trying to compete in a tough medical arena. 173

187 The Role of Healthcare Professionals in the Diffusion of Patient Innovations: An Experiment in the Field of Physical Medicine and Rehabilitation Poster Merle Schlottmann (Católica-Lisbon School of Business and Economics), Pedro Oliveira (Católica-Lisbon School of Business and Economics), Helena Canhão (Nova Medical School, Lisbon) Research has shown that both patients and their informal caregivers are a powerful source of innovation (e.g. Oliveira et al., 2015; Canhão et al., 2016). Using the opportunities provided by the internet, patients become increasingly informed about healthcare and change their role from a passive consumer of care ( ) to active partners in the healthcare process (Amann & Rubinelli, 2016). Patient innovators (including patients, caregivers and collaborators) start to develop solutions for their unsatisfied needs, as there are no products or services available or affordable on the market. Even though it has been shown that numerous patient innovations exist (e.g. Oliveira et al., 2015), their diffusion is still rare, since the innovators did not intend to share their solutions in the beginning. The open platform Patient Innovation enhances diffusion by encouraging patients to share and search for patient-driven solutions. Apart from low levels of diffusion, the role of healthcare professionals in the patient innovation development and diffusion process is understudied. Sometimes different levels of risk taking willingness is a reason for patients to start developing solutions on their own without integrating their physicians (Habicht, Oliveira & Shcherbatiuk, 2013) and they step aside from the regulated innovation process (Flowers, 2014). In addition, mistrust in healthcare professionals and medicine leads to adapt patient innovations (Zejnilovic et al., 2016). Further, Oliveira et al. (2015) raise the question of a principal agent problem between patients and physicians as they found out that there exist different opinions on patients needs as well as on the usefulness of solutions. Nevertheless, research shows that the integration of patients in the process of delivering healthcare service by multidisciplinary healthcare teams fostered the 174

188 innovative behaviour of patients in number and variety (Hannemann, Weber- Schulz, 2014). In the end, physicians are crucial in the process of evaluating the safety of the solutions. Aware of the lack of diffusion of patient innovations and the increasing gap between patients and physicians, we aim to examine the role of healthcare professionals in the diffusion process. Therefore, we chose a qualitative multiple case study approach and conduct semi-structures interviews and surveys with patients and their healthcare professionals (physicians as well as physical and occupational therapists). The sample includes commercialized and non-commercialized solutions in the field of physical medicine and rehabilitation. We examine if solutions were accepted as advisable useful treatment by the physicians and further, if and to which degree physicians were informed about the solution and took part in the diffusion of patient innovations. Currently, six interviews have been conducted with innovators, being two caregivers and four patients. So far, only one treating therapist has been interviewed and four more healthcare professionals agreed to participate in the survey because either the healthcare professionals are already retired, have not answered or they were not named by the respective innovator. After the first interviews, we noticed that therapists were involved to a higher degree than physicians in both the development and diffusion. Patients explained this with the fact that the number of appointments with therapists is higher than with physicians. The different approaches between physicians and therapists are seen by patients as the reason for therapists to be more likely to adapt experimental solutions. In addition, patients pointed out causes such as lack of time, responsibility, liability, and absence of medical approval as main barriers to motivate their physicians to contribute to the diffusion of the solutions. We have observed that patients rarely ask their physicians for help in terms of the development of the solution. But as soon as they want to share their solutions with others, either for reasons of testing or for commercialising, innovators see their physicians and therapists as a promising channel to reach a broader network and increasingly ask them for support. But in four out of six cases, physicians did not contribute to diffusion activities. In contrast to participation in the development, the majority of innovators assume that healthcare professionals could increase their sharing activities. There is a repeating perception among the innovators that participation of healthcare professionals increased after the media became interested and first tests have been 175

189 successfully reported. Based on these first results of the multiple case study that suggested the media as promising channel to enhance the diffusion of patient innovations, we aim to examine if actively integrating physicians by usage of media can confirm this perception and increase the acceptance of patient innovations among physicians. Therefore, we run an experiment in the field of physical medicine and rehabilitation presented on the Patient Innovation platform. Approximately 100 physicians and residents in physical medicine and rehabilitation are provided with three newsletters over a short period of two months, each presenting three patient innovations from their specialty. We evaluate two distinct groups of solutions: those featured vs. not featured in the newsletter and compare the quantity of solution-related feedback (views, likes, comments). The development of the level of acceptance of patient innovations among physicians will be evaluated through a survey which is going to be sent along with the newsletter with the physicians. For this purpose, acceptance will be measured by their intentions to share solutions with other patients. 176

190 How Lead Users Impact the Decision-Making in Innovation Process? Poster Senda Belkhouja (Grenoble Ecole de Management), Corine Genet (Grenoble Ecole de Management), Vincent Mangematin (Grenoble Ecole de Management) As defined by Von Hippel, lead users have two characteristics: First, they have high incentive to solve problems. Second, what lead users want in the present is what the market needs in the future. Lead users may be mobilized as a source of innovation and ideas in the innovation process as well as the diffusion process (Von Hippel 1986, 2005, and 2007). Involving lead users in the innovation process seems to be particularly efficient at the beginning of the product life cycle (Aaron et al., 2012; Lettl, 2006) in radical innovation projects rather than in incremental ones (Chatterji and Fabrizion, 2014). Besides being external knowledge sources, lead users can also play a role to play in the innovation process within firms (Hienerth and Lettl, 2017; Katila, 2017). Comparing external lead users to internal lead users and ordinary employees in terms of innovation, Schweinfurt (2017) finds that internal lead users generate higher quality ideas compared to ordinary employees, but, less original ideas than external users. Internal users ideas remain more feasible than those of external lead users because of their awareness of the organization s internal implementation procedures. While the literature has provided evidence on the role of internal and external lead users ideas and knowledge on the innovative activity of the firm, it has mostly neglected the role that lead users play in decision-making within the firms. Prior literature shows that lead users in young firms of medical devices, are more innovative in technical and governance roles than in executive roles (Katila, 2017). This result seems to hint at that users can be helpful by expanding the variety of solutions to the firm s innovation problems, but are significantly less helpful in just improving their selection. While prior literature suggests that lead users influence the form in which decision-making is taken within the firm, by the nature of their empirical data and analysis, they cannot say much about how lead users influence the process of decision-making within the firm. Hence in this study, we will attempt to address the following research question: What role do lead users play in the 177

191 decision making process of innovation projects? To examine empirically this issue, we will rely on four projects aimed at new product development within one biotech company specialized in diagnosis: One project is incremental, three others are radical. Three of these projects include lead users in their project. Data from these four projects will be collected through interviews with different internal employees (Engineering, Biologists, Marketing) and secondary data such as technical documents (ex: project documents, patents), meeting reports, and documents of internal communication. Relying on this data, we will examine influence of lead users on the decision to pursue the project throughout the project life cycle (i.e. from ideas generation to product development). First, we will compare the standard innovation process of the firm to the real process of each project in order to identify the differences and check whether these differences are the same from one project to another. Second, we will study the role of each actors in each project and the impact of each actors (internal and external) on the decision-makinf process of innovation (lead users, employees, marketing department, decision makers ). Results suggest that innovative projects that include lead users are more likely to reach the product development phase. This seems to result from the fact that decision makers perceive the participation of lead users in the project not only as a source of innovation, but also as an additional warrantie that project has higher value on the market. 178

192 The Impact of User Innovation on Patients Health-Related Quality of Life An Explorative Case Study on Medical App Developers Poster Moritz Göldner (Hamburg University of Technology), Cornelius Herstatt (Hamburg University of Technology) Prior research has shown that users are a valuable source of innovation in the healthcare sector. There is evidence that patients and their caregivers develop medical devices for their own need (Goeldner & Herstatt, 2016). However, little is known on the impact of the development on the wellbeing of the patient. According to WHO, "Health is a state of complete physical, mental, and social well-being, and not merely the absence of disease or infirmity." Drawing on this definition, we use the construct of health-related quality of life to asses if the development and the usage of the self-developed medical device has an influence on the patient or the caregiver. First studies on the assessment of change in quality of life of user innovators yielded positive results (Oliveira, Zejnilovic, Canhão, & von Hippel, 2015). However, measurement of healthrelated quality of life and associated measures is a complex task that needs to be improved. Further, as an increase in quality of life is closely related to an increase in well-being and life satisfaction (Steptoe, Deaton, & Stone, 2015), further studies on the overall health status are necessary to confirm those initial findings. Building on this previous work, our study aims to quantify if and if yes, how the health-related quality of life is influenced by the development of a medical device and its usage. Our empirical field is the markets for medical smartphone applications, where both, patients and caregivers are developing and publishing their self-developed apps next to professional companies. From a regulatory perspective, medical smartphone apps might be classified as a medical device according to European or U.S. laws. In this explorative study, we conducted 10 semi-structured interviews with medical smartphone app developers (seven patients and three caregivers) from Germany, Austria, Finland, UK, Switzerland and the U.S. We aim at finding out if their health-related quality of life changed after the development medical app has been completed. 179

193 Further, we assess if the health status has improved as well. We have two major findings: First, we find that all apps were developed because no proper alternative was in place at that time. Thus, all apps were novel with respect to at least one important feature like visualization, complexity or design. Some apps just combined existing data for instance the developer of a migraine treatment app: Building on the experiences with his flat mate, he designed an app that uses available data on meteorological conditions (atmospheric pressure and temperature) and pollen count data and combined it with the GPS-position of the user. This unique combination of data helped the flat mate to better understand the patterns of migraine and to improve the health outcome. All apps developed from the interviewees target chronic conditions such as hypertension, diabetes or migraine. The data-driven medical apps provide useful information to both, the healthcare professional and the patient. Second, we find that in all cases the usage of the self-developed medical app enabled patients a better self-management of their disease and consequently led to an improvement in health outcome compared to the time before the app development. By constantly tracking their data, the patients recognized patterns and were able to change those patterns accordingly. By using the app, I learned what causes higher blood pressure and what not. So I adapted my way of living and it really helped me (Patient 1) A diabetic who developed an app according to his own needs was able to improve his health outcome: It had a significant impact on the HbA1c-value and my overall wellbeing (Patient 6) Our interviewees confirmed the direct connection between the health outcome and their health-related quality of life. Particularly the caregivers, who developed the app for somebody, mentioned the positive impact on the well-being of the user of the app. In some cases, their own quality of life has increased: particularly if the app was developed for a close relative, the positive influence of the app usage on the patient impacted the developer as well. The overall findings are in line with a current literature stream that indicates that patients are gaining more and more influence on their treatment, are taking earlier actions in order to increase their quality of life and subsequently improve the health outcome of their peers. Firms in the healthcare sector should take advantage of this and consider including patients and caregivers in their development process. Next, we will develop a questionnaire for a larger quantitative survey to confirm these initial results. 180

194 Patientube - Supporting Patients and their Relatives by Curated Expert and User Generated Peer Videos An Analysis of Critical Success Factors Poster Andrea Hofmann-Rinderknecht (Patientube AG), Andreas Kreimaier (Patientube AG), Gerhard Buchegger (Patientube AG), Dominik Walcher (Patientube AG) Quite often a severe medical diagnosis represents a shock and existential crisis for the individual herself as well as her relatives. Uncertainty, stress and desperation are common consequences, which regularly manifests in isolation and feeling of loneliness. In addition to doctors comrades in misfortune can give support and relief quite often even more trustworthy due to the fact they have experienced the same. Swiss startup Patientube ( is an online platform offering professional expert videos with medical doctors and user generated report clips produced by peers (i.e. patients as well as relatives). Persons concerned have the possibility to find appropriate clips addressing their needs. The platform s mission is transforming alone into unalone! Patientube has emerged from the Swiss Patient Forum and can be seen as expert and peer counseling platform representing a self-help group in a timely format (rf. Youtube). The platform is supported by several institutions (e.g. interest groups, insurances etc.) and in beta stage. At the moment all contributions are curated. Besides offering more expert videos a wide rollout with reaching the critical mass for a vital sharing and self-regulating dynamic (i.e. content is consumed, generated and monitored by users) is the objective. To accompany these activities a research study is currently conducted with Salzburg University of Applied Sciences (DERESA Center for Co-Creation at the Department of Design and Product Management). With the help of qualitative interviews, online surveys and experiments the answers and reactions of about 50 participants (patients and relatives) are analyzed in order to identify the benefits and critical incidents of the platform and the presented videos. Within three polls different questions are investigated. Among others it is asked what is the ideal length of a clip, what topics should be addressed and which depth and level of detail is most 181

195 helpful for the visitors. Till 1 July 2017 the empirical part of the study is completed, comprehensive data is expected to be gathered. We will present the central findings and the derived implications for the Patientube platform. 182

196 Sustainable Healthcare: Mechanisms of Co-Innovation by Patients and Providers Poster Adam Seymour (ICN Business School) Researchers generally agree of the patient's role as source of innovation in healthcare industry. However, relatively little research examines innovation capacity of patients and providers, and their ability to develop new products and services. Prior research in this area suggests that patients have been viewed as having a relatively passive role, essentially a recipient of what an organization does for them (Payne et al., 2008). This view has been prevalent in healthcare (Berry & Bendapudi, 2007; Lorig & Holman, 2003). This study examines some of the positive and negative impacts of patient innovation, and examines the role of the patient in shaping healthcare delivery systems of the future. The objectives in this study are to explore: In what ways can patients contribute to the design of new products and services? How should healthcare organizations be re-designed to better support co-creative practice principles? What are the characteristics of patients who are motivated to take on high levels of responsibility? How can more sense be made of collected patient s innovations for patient population? A growing stream of literature argues for a more active role for patients in healthcare delivery (Hibbard, 2004; Wallerstein,2006; McAllister, 2012; Dyson, 2009; Swan,2012; Frydman, 2009; Ferguson, 2016). However, this literature does not consider the innovation capacity of patients and their ability to design and develop new products and services. To attain the goal of truly patient-centric healthcare, it is necessary to also support and integrate the innovation-related activities of patients and providers into the current system. For example, Innovation programs can provide this intermediation through multiple mechanisms such as innovation fellowships; competitive pilot grant programs; and efforts directed at soliciting the ideas of providers, such as ideation sessions (short meetings with a group of providers in one department, in which they pitch their ideas to the innovation staff), hackathons (events that occur over a weekend and bring together providers, 183

197 engineers, and programmers to design innovations), and sandpit events (residential interactive workshops that spans multiple days to drive radical approaches), (Bates, Sheikh, & Asch, 2017). The engagement and the close cooperation with treatments were demonstrated to be more effective with enlightened and instructed patients. Particularly in hospitals, more than a few patients felt more comfortable to share their experience with nurses, whose relationship is more likely to be characterized by trust and equality (European Commission, 2012, p. 37). Moreover, research shows that significant numbers of patients want to interact and co-create value, and not necessarily with just one service provider but also with communities of professionals, service providers, and other customers (McColl-Kennedy et al., 2012). Besides, the involvement is expected to be more intensive with chronically ill patients where, the percentage of patient developed treatments and devices varies across the diseases, namely in: 42% for Cystic Fibrosis, 50% for Asthma, 50% for Sleep Apnea, 47% for Cancer, and 79% for Diabetes (Shcherbatiuk & Oliveira, 2012). The value co-creation is part of a larger theoretical framework based on the understanding that value emerges in context when resources are integrated. Changing the view of the patient from passive to active is a shift from goodsdominant logic toward service-dominant logic (Vargo & Lusch, 2004; 2008). In addition, without a better understanding of the effects of customer involvement on product outcomes, it is unclear whether such efforts can lead to successful product innovation. The conceptualization of value co-creation activities in this paper suggests a need for healthcare service providers to adopt approaches such as New Product Development (NPD), Theory of Planned Behavior (TPB) and Theory of Reasoned Action (TRA) that would effectively integrate patient resources to co-create value. The research design is exploratory and qualitative, due to the rather new focus on the interaction of patients and providers with the system context, and the current lack of available knowledge on this topic in the light of the sustainability challenge in healthcare. Inductively analyzing the data is most valuable for gaining new valuable insights for both innovation theory and practice (Yin,1994). 184

198 Sharing Economy and Platforms 185

199 Explaining the Vertical to Horizontal Transition in the Computer Industry Paper Carliss Y. Baldwin (Harvard Business School) This paper seeks to explain the technological forces that led to the rise of vertically integrated corporations in the late 19th Century and the opposing forces that led to a vertical-to-horizontal transition in the computer industry one hundred years later. I first model the technology of step processes with bottlenecks and show how this technology rewards vertical integration, a hierarchical organization, and the use of direct authority. These properties in turn became the organizational hallmarks of so-called modern corporations. I then model platform systems, showing that, in contrast to step processes, this technology rewards the multiplication of options, increasing risk, and modularity. Moreover, given a modular architecture, a platform system can be open, with different components supplied by separate firms with no loss of interoperability or efficiency. Openness multiplies options and expands diversity, thus increasing the platform system s value. The last two decades of the 20th Century saw the rise of three distinct types of open platforms in the computer industry: (1) forward open platforms with downstream complementors; (2) backward open modular supply networks; and (3) open exchange platforms designed to facilitate transactions and other forms of social interaction. Whereas in 1980, vertically integrated firms dominated the industry, by 2000, the verticals had essentially disappeared. The largest firms in the industry in 2000 were sponsors and participants in open platform systems. I argue that the vertical-to-horizontal transition in the computer industry was an organizational response to a fundamental change in economic rewards to the technologies of rationalized step processes vs. open platform systems. 186

200 Platform Ecosystems: How Developers Invert the Firm Paper Geoffrey Parker (Dartmouth College), Marshall Van Alstyne (Boston University), Xiaoyue Jiang (Boston University) For a period starting in 2015, Apple, Google, and Microsoft became the most valuable companies in the world. Each was marked by an external developer ecosystem. Anecdotally, at least, developers matter. Using a formal model of code spillovers, we show how a rising number of developers can invert the firm. That is, firms will choose to innovate using open external contracts in preference to closed vertical integration. The locus of value creation moves from inside the firm to outside. Distinct from physical goods, digital goods afford firms the chance to optimize spillovers. Further, firms that pursue high risk innovations with more developers can be more profitable than firms that pursue low risk innovations with fewer developers. More developers give platform firms more chances at success. Our contribution is to show why developers might cause a shift in organizational form and to provide a theory of how platform firms optimize their own intellectual property regimes in order to maximize growth. We use stylized facts from multiple platform firms to illustrate our theory and results. 187

201 Who Should Benefit in a Sharing Economy Model: A Look at Resource Endowment and Trust Paper Christopher Smolka (WHU - Otto Beisheim School of Management), Christoph Hienerth (WHU - Otto Beisheim School of Management), Franz W. Kellermanns (Belk College of Business) The recent emergence of sharing economy models has established novel organizational forms which are founded upon individuals who participate in commonly sharing resources. In this paper, we seek to understand the factors that determine individuals expectations towards beneficiaries in sharing economy models. Utilizing survey data from the organization Gawad Kalinga, we investigate how the resources an individual is endowed with interact with trust at different levels (on the organizational, community and institutional level). Implications for research and practice are discussed. 188

202 To share or not to share Exploring the Impact of Sharing Behaviour on User Innovativeness Paper Frank Tietze (University of Cambridge), Thorsten Pieper (Hamburg University of Technology), Carsten Schultz (Kiel University), Cornelius Herstatt (Hamburg University of Technology) We observe a rise of new business models that embed different forms of tangible product sharing. Innovation research and particularly user innovation research has hardly considered this topic. This paper sets out to conceptually and empirically explore the impact of users product sharing behaviour on users innovation behaviour. We contribute (i) a new concept labelled sharing experience and its operationalization based on the well-established use experience construct, (ii) a typology to categorize sharing types and (iii) a first empirical analysis quantifying the impact of sharing on user innovativeness. Primary data was collected from a large German farmer sharing community established since 50+ years. The survey yielded 1,064 responses. Our results show that sharing has a significant positive impact on the user s innovation behaviour. Users technical expertise positively moderates this relationship. The paper proposes four sharing types based on the two dimensions in- and out-sharing revealing significant differences across these types. 189

203 Open Innovation in the Digital Age. New Options to Close the Gap between Universities and Companies? Paper Boris Alexander Becker (FernUniversität in Hagen) The image of the university as an ivory tower is well known and frequently discussed. The question rising is what picture holds true today for the university as a stakeholder in society. One point of interests marks the openness of universities towards other stakeholders in society. In the following paper the focus lies on the topic of Open Innovation, which considers this new demand of openness for universities with regard to the collaboration with private companies. Under the item of Open Innovation universities become an important stakeholder within the imperative to innovate. In fact the collaboration of universities and enterprises is judged to be crucial to trigger innovations and to secure a socio-economic development. Within this discussible paradigm the paper focuses on digitization as a tool to close the gap between the different stakeholders on a boundary or local spanning level. For this purpose first of all the general state of the art with regard to university-industry collaboration and current options and detriments are discussed. Thereby the thesis of the stickiness of implicit knowledge transfer to personal mobility is put into question. Based on this assumption the idea of digital tools and especially digital platforms as an intermediate is elaborated to circumvent the locality of knowledge transfer and to boost knowledge circulation. Until today these digital tools are more widespread within business to business cooperation. The question arises in which way they are useful to facilitate knowledge transfer between universities and industry. Trust, intellectual property rights and opportunistic behaviour are critical aspects for the collaboration between university and industry and are often conceptually approached by transaction cost- as well as network theory. To the best of the author s knowledge this is done mainly with regard to business to business cooperation and should be adapted to the reality of universities. Thereby digitization as a possible intervening variable is introduced. Referring to the possibilities of digital tools 190

204 with regard to trust and network effects, the paper considers their suitability to mitigate the gap between scientists and entrepreneurs. Summing up mainly theoretical considerations lead to four leading hypotheses with regard to innovation outcome within the frame of digital environments to overcome the gap between scientists and entrepreneurs. This may be the basis for further empirical exploration. 191

205 Principles of Crowd-Based Organizing: Unpacking Platforms Functions and Processes Poster Robert M. Bauer (Johannes Kepler University Linz), Thomas Gegenhuber (Johannes Kepler University Linz) Platforms provide essential infrastructure for crowd-based organizing (Powell, 2016). They serve as organizational backbones for distributed innovation phenomena including digital user innovation, crowdsourcing, open innovation (Afuah & Tucci, 2012; Baldwin & von Hippel, 2011; Dahlander & Gann, 2010; Jeppesen & Frederikson, 2006; Jeppesen & Lakhani, 2010; West & Bogers, 2014) and, on the other, distributed execution of routine tasks including click work or services provided by the so-called sharing economy (Bauer & Gegenhuber, 2015; Davis, 2016). We understand platforms as organizational entities that, first, identify and mobilize external sources (e.g. individuals, teams or organizations) via broadcast calls and, second, provide means to organize the contributions from these sources in order to reach a certain goal (i.e., solving a problem or executing a task). Crowd-based platforms emerge in a bewildering variety: Innocentive brings together organizations with an expert crowd to solve innovation problems; MechanicalTurk is a digital assembly line for globally distributed execution of microtasks; AirBnB allows travelers around the world to book homes instead of hotels; Leedir presents the 21st century version of the wanted poster ; Recaptcha hijacks websites sign-up systems to transcribe books and Google leverages user patterns for selling online advertising. Despite their utterly different forms and appearances, they all rely on broadcast search on explicit or implicit calls for external sources that are willing and capable of providing products and services in exchange for financial or non-pecuniary benefits (Bauer & Gegenhuber, 2015). We agree with Kornberger (2016) that a new language for describing and explaining these phenomena is needed. Various typologies and frameworks in the literature capture how these platforms operate (e.g. Malone et al., 2010). This prior work has immensely propelled our understanding of platforms, yet it is still quite limited sometimes 192

206 oversimplifying platforms by zeroing in on one dominant governance mode at the expense of critically important secondary modes (e.g. Wikipedia is generally dubbed as collaborative, yet it is also highly selective); neglecting crucial elements in crowd-based organizing, such as the type of broadcast call (Malone et al., 2010); or exhibiting excessive levels of analysis such as highly abstract frameworks, not (yet) operationalized for empirical purposes (Kornberger, 2016), or rather specific ones that hardly apply across platform types (e.g. Geiger et al., 2011). In this conceptual piece, we aim to fill this gap by developing a general framework, denoting the functions and processes essential to all platforms engaged in crowd-based organizing. We maintain that implicitly or explicitly all platforms draw together three aspects: First, the informational aspect includes the broadcast call, the subsequent conveying and processing of responders contributions, and a concluding review, in which the involved parties (i.e. demand, supply, and platform) assess their satisfaction. Second, the transactional aspect addresses the exchange of rights for financial or nonpecuniary benefits. Third, the socio-material aspect addresses the actual carrying out of work resulting in the production and delivery of digital or physical objects, or the provision of services. Specifically, our general framework comprises six functions, each with a distinct goal. To achieve these goals, each function relies on several processes that capture, to put it rather crudely, what platforms actually do: (1) Call: Broadcast calls aim at commitment (i.e. decision to participate and commit resources) from all members of a target group. Platforms specify target groups and mobilize contributors through calls the target group attends to, understands, and values (Afuah & Tucci, 2012; Bauer & Gegenhuber, 2015; Jeppesen & Lakhani, 2010). (2) Convey: The aim of conveying is availability of maximum relevant input. The platform must thus define (the standard for) acceptable input, admit accordingly produced input, provide a channel to transmit the contribution, and collect the received data in an organized manner (Brabham, 2013; Doan et al., 2011). (3) Process: Processing received input is aimed at capturing its relevance and value. Platforms enable compression of received contributions into a single solution (i.e. task completion) through (a) selection, i.e. choosing the best according to various human- and machine evaluators and elaborate evaluative procedures (e.g. multiple party and multi-stage assessments); (b) aggregation, i.e. adding up all contributions to benefit from their total effect; (c) integration, i.e. a creative 193

207 combination of selection and aggregation (Bauer & Gegenhuber, 2015; Kornberger, 2016). Some platforms enable (re-)iterations of from creating a contribution to conveying and processing. (4) Transact: Transacting aims at enforceable agreements on deals between the demand and supply side (Howells, 2006; Kohler, 2015). Platforms facilitate mutual agreement on transferring rights in exchange for economic gains (i.e. money), options for future gains (e.g. learning or reputation), or non-pecuniary gains (e.g. fun, recognition or identification/value; Füller, 2010; Lakhani & Wolf, 2005; Franzoni & Sauermann, 2013; Shah, 2006). Platforms further guarantee secure transfer of rights and resources within an institutional and legal framework (Howells, 2006; Morgan & Wang, 2010). To minimize transaction costs, platforms can standardize communication (about contributions) by, for instance, offering price categories (Moldoveanu & Bauer, 2004; Martinez & Dacin, 1999). (5) Execute: Execution refers to the creation and/or delivery of digital objects or physical products, or provision of services. Platforms shape execution: they (in-)directly enable contributors (e.g. training); they organize and lead them (i.e. categorizing the crowd based on competencies, prior successes or integrity and monitoring contributors); and they step in, carrying out certain steps in the production chain or enlist third parties to do so. (6) Review: Reviewing aims at (public) evaluation of completed transactions. Assessments of benefits and convenience can be made public through unilateral or bilateral rating, and commenting. Such evaluations typically serve as input signals for future platform interactions (Malone et al., 2010; Orlikowski & Scott, 2014). In this extended abstract, we have suggested a repertoire of what platforms do. This vocabulary allows us to distinguish platform types based on which functions and processes are (de-)emphasized and on the specific sequence in which they occur. In addition, we seek to put forward a full range model that also specifies the processes involved in mobilizing and qualifying demand and supply. 194

208 The Free Strategy : Economics of Open Versus Proprietary Designs Poster Alfonso Gambardella (Bocconi University), Eric von Hippel (MIT Sloan School of Management) In this poster session, I explain why proprietary products and services purchased by industrial customers today will increasing be displaced by substitutes based upon open source designs. These open source designs will be developed by input purchasers themselves. The trend towards free we discuss and model is being driven by exogenous shifts in technology, prominently including the transition to digital design, and the transition to communication via the internet. The net effect of these shifts is to make a free innovation strategy increasingly profitable for input purchasers relative to the purchase of proprietary inputs developed by monopoly suppliers. The shift from protected to open source designs also represents a novel form of challenge to the patent system: one involving potential customers increasingly walking away. Successful implementations of the free strategy are visible today among groups of both large and small input purchasers. Thus, in the Open Compute Project, major users of web server hardware like Facebook and Google have joined with others to collaboratively create and freely reveal open hardware designs for products purchased in massive numbers for web service data centers. These open white box designs are competing very successfully and profitably - against the proprietary designs offered by traditional suppliers. At a smaller scale, scientists are increasingly creating and openly sharing designs for scientific instruments that they use as inputs for their work via a grass-roots effort. Copies of these open source designs are self-produced by the scientists at major savings often over 90% - relative to purchases of substitute proprietary designs offered by scientific instrument manufacturers. 195

209 Pay-As-You-Drive Models in The Sharing Economy: A Comparison of German and U.S. Car Owners Poster Mario Schaarschmidt (University of Koblenz-Landau), Raoul Könsgen (University of Koblenz-Landau), Björn Höber (University of Koblenz-Landau), Patrick Hacker (Thyssen Krupp) Sharing physical goods or unused spots has become a prominent trend in recent years and involves sharing rooms (AirBnB), tools (Snap Goods), and cars, either as an adhoc service (Uber) or more goods-oriented (Wheelz) (see Hamari et al. 2015; Malhotra and Van Alstyne 2015). However, sharing involves risks on multiple sides, that is, the sharing offeror, the user, and most likely a matching platform as in the case of Uber or AirBnB (Evans 2012). While the platform operator usually offloads risks to others (Lanier 2013), as an owner of a car, for example, one might fear misbehavior by potential users / customers. In turn, the customer might fear bad experiences or services because single sharing offerors have limited options to signal their trustworthiness to potential customers. Much of the risk inherent in sharing is addressed with online reputation systems, such that customers can rate the offeror (and vice versa) to decrease information asymmetries (Dellarocas 2003). For car sharing especially, technology-supported pay-per-use options might reduce inherent risks. In particular, pay-as-you-drive models use technology to monitor driving behavior that can act as a basis for lowering or increasing prices for usage (Desyllas and Sako 2013). This approach has been first applied in car insurance settings and first studies acknowledge their effect on driving behavior. For example, Bolderdijk et al. (2011) showed that the introduction of a pay-as-you-drive insurance fee significantly reduced speed violations of young drivers. Given the potential of pay-as-you-drive models to reduce risks, they are suitable instruments to reduce risks in car sharing scenarios. However, knowledge concerning the effectiveness of such instruments remains scarce. For instance, personal attachment to a car might be an important predictor of sharing behavior (Nilsson and Kueller 2000), but these attitudes have not been investigated in the context of the sharing economy. 196

210 Moreover, cultural differences may exist in sharing cars in particular. To this end, we conducted an initial exploratory survey in Germany (N=240) and the United States (N=751). The survey contained questions abou car usage and demographics. It is followed by a detailed description how pay-as-you-drive could work for private car rental / sharing. For both countries, we relied on crowdwork-based platforms to recruit participants (Clickworker for Germany and Amazon Mechanical Turks for U.S.), thus, respondents were paid for participation. We excluded all participants that indicated to not own a car. For the German sample, 109 (45.4%) respondents were female and the mean age was For the U.S. sample 323 (42.8%) were female and the mean age was Thus, both samples are comparable in terms of gender and age. Our preliminary results indicate that U.S. residents report higher willingness to use pay-per-use payment options. In addition, U.S. based residents have more confidence in the effectiveness of such systems (Answer to the question = I think that customers will drive more gently with my car when they know that I can track their driving behavior. MGermany=2.31 vs. MUS=3.21; Table 1). To the actual state, the survey yield first insights into an underexplored research area in the context of the sharing economy. The next steps involve investigating causal relationships and cultural differences. The final findings will help car manufacturers to design potential pay-as-you-drive-monitoring solutions aligned to the needs of specific target markets. 197

211 The Role of Users in the Platform Economy A Multiple Case Study of Institutional Change by Users of Airbnb Poster Wouter Boon (Utrecht University, Innovation Studies Group), Kristy Spruit (Utrecht University, Innovation Studies Group) Peer-2-peer [P2P] service platforms have seen an increase in numbers over the last few years and have displaced, altered, or threatened existing and often regulated markets. This has boosted discussions about the economic value of P2P service platforms and their impact on traditional businesses with which these platforms compete. Moreover, these platforms not only initiated economic debates, also the social impact is part of the overall discussion, e.g. labour exploitation, unequal access for low-income and minority communities, and the status of regulation and taxation. Contradictory to these discussions about legitimacy and legality, P2P service platforms have seen an enormous growth in number of users in their first few years. The enormous growth contributes to gaining legitimacy, legality and momentum, which has been discussed to be important for successful platforms as often the first to succeed also gains the best market position; the winner takes all. This paper focuses on this phenomenon in particular for P2P home sharing platforms, as these are prone to all economic, social, legitimacy and legal topics of discussion and have seen different developments in different urban institutional contexts. A definition of home sharing platforms similar to the definition of platforms of Busch et al. is used, where platforms enable customers, be it a consumer or not, to order or provide access to a home in exchange for payment. In order for these home sharing platforms to become part of the formal economy and institutional frameworks, institutional change is needed. For this transition to happen, platforms have to overcome the discussion on legality and legitimacy, and institutions have to be created, maintained or disrupted. Institutional change has had much attention of scholars and studies have been performed into the activities that have to be executed to obtain institutional change. In these studies, focus lies on the individual actor or the organisation that function as an 198

212 institutional entrepreneur from within or from the outside of a field. However, platforms work differently than traditional businesses and have many heterogeneous users that have to be attracted and engaged, and that can execute activities on behalf of the platform. Users are understood as crucial to a platform s capacity to cultivate and capture value as they are not solely consumers but they are the creators, producers, and generators of value, content and data. The role of the consumer has thus changed and become more important. However, not much is known about users of sharing platforms. Previous research has studied users of the internet and of social media. Groups of users have been categorized in order to structure these divergent and heterogeneous users. These studies focused on what drives users to use or not use, for example the Internet, while the exact activities and how users can contribute to the cultivation and captivation of a platform s value has not yet been studied. As users have a more proactive role because of these platforms, it is expected that different user groups can facilitate or oppose the institutionalisation of a platform. However, the reason why certain user groups choose certain activities to change institutions in an urban context and why some succeed and others do not, has also not yet been studied. This paper therefore aims to extend existing theory by conceptualizing what characteristics and conditions are relevant in the case of institutional entrepreneurship by users of a platform. The aim is to provide a better view on what different user groups exist, what activities they perform, why they perform these activities and if and how these activities of different user groups contribute to institutional change. Therefore, the main question of this study is the following: What is the role of users of home sharing platforms in influencing institutional change? The question will be explored during a multiple case study of three cities in which home sharing platform Airbnb is active; New York [USA], London [UK], and Amsterdam [NL]. Airbnb is chosen as a platform as this is currently the largest home sharing platform. It has seen many debates about its legitimacy and legality and is therefore interesting to study in the context of institutional change. New York, London, and Amsterdam are chosen as cities as Airbnb has seen different institutional outcomes in these three cities. In London and in Amsterdam Airbnb is part of the institutional setting as it is allowed, however in both cases under different circumstances. In New York Airbnb regulation is very strict and states one host, one home. In all three cases Airbnb has seen a 199

213 different development heading for the formal economy. Data is collected from four different data sources: 1) in-depth semi-structured interviews with policymakers, users, and other experts of the municipality of Amsterdam, New York, and London; 2) media reports about Airbnb in newspapers; 3) user forum data; and 4) communication between Airbnb and its hosts. A process approach enables the mapping of these activities and events, which subsequently explains change. For this step, this study makes use of event history analysis. 200

214 Author Index Von Krogh G. 2, 12 Conradie P. 168 Al-Dulaigan A. 33 Danler M. 83 Al-Jayyousi O. 33 De Jong J. 147 Albarran N. 170 De Marez L. 168 Alexy O. 92 Deichmann D. 27 Antons D. 34 Deimel M. 63 Apel S. 6 DeMenna M. 170 Azevedo S. 169 Dimitrova S. 42 Baccarella C. V. 88 Ebersberger B. 30 Baldwin C. Y. 186 Ehls D. 143 Banken V. 57 El Mezouaghi R. 56 Bauer R. M. 26, 192 Enell-Nilsson M. 18 Becker B. A. 190 Evgeniia F. 135 Belbaly N. 56 Fecher F. 41 Belkhouja S. 177 Feldmann N. 86 Ben-Menahem S. 12 Figge P. 70 Benker A. 107 Fisher G. 146 Benz C. 86 Franke N. 15, 52, 101, 109, 129, 135, 144 Bereczki I. 75 Frederiksen L. 27, 82 Beretta M. 27 Füller J. 29, 41, 80, 128 Berreiter S. 6 Fursov K. 142 Bertram M. 158 Galia F. 30 Bhattacherjee A. 29 Gambardella A. 195 Bianchini S. 96 Garaus C. 49 Blasco A. 48 Gatzweiler A. 43 Blazevic V. 43 Gault F. 122 Boon W. 198 Gegenhuber T. 26, 192 Bortoluzzi G. 104 Genet C. 177 Brandonjic P. 144 Ghasemzadeh K. 104 Buchegger G. 181 Göldner M. 95, 179 Bugawa A. 33 Gomez-Marquez J. 170 Buschmann A. 60 Greul A. 55 Calin Gurau 56 Griffith T. 92 Canhao H. 169 Hacker P. 196 Chappin M. M. H. 68 Haefliger S. 2 Chaudhary N. 48 Halbinger M. 3 Chen J. 93 Hanisch M. 6

215 Hannen J. 34 Kuusisto J. 124 Hanson S. 154 Lakhani K. R. 48 Häussler C. 6, 70 Laursen K. 30 Heald S. 134 Lee W. 141 Heimeriks G. J. 68 Lendowski E. 60 Heiss M. 128 Lettl C. 63 Heite J. 37 Liang L. 124 Henkel J. 8 Lichtner R. 66 Hepp D. 8 Linton J. 142 Hermann A. M. 68 Llerena P. 96 Herregodts A. 168 Luethje C. 144 Herstatt C. 31, 95, 143, Lukoschek C. S , 189 Heuschneider S. 143 Maier R. 57 Hienerth C. 188 Maillart T. 2 Hilgers D. 121 Majchrzak A. 92 Hirota A. 100 Malonis P. 134 Höber B. 158, 196 Mangematin V. 177 Hofmann-Rinderknecht A. 181 Marshall D. 170 Hoisl K. 37 Mehner B. 101 Holthaus C. 131, 159 Melillo F. 3 Homscheid D. 158 Menietti M. 48 Hu P. 134 Merz A. 57 Huang J. 93 Metz F. 129 Hutter K. 41, 80, 128 Moors E. H. M. 9 Jaskari M. 18 Moritz M. 76 Jiang L. 154 Moschner S. L. 31 Jiang X. 187 Mujika-Alberdi A. 145 Kantola J. 18 Nobis M. 109 Karlusch A. 163 Nusbaum H. 134 Keinz P. 109 O'Hern M. 94, 154 Kellermanns F. W. 188 Oliveira P. 169, 171, 174 Kim Y. 141 Olk P. 119 Kleinscheck R. 80 Otitigbe J. 137 Kohler T. 28 Page M. 129 Koller H. 7, 132 Parker G. 187 Könsgen R. 196 Patsali S. 96 Korbel J. J. 4, 120 Pavlova K. 39 Krämer K. 39 Pescher C. 128 Kreimaier A. 181 Pieper T. 189 Kruse D. 95 Piller F. 34, 43, 97

216 Pohlisch J. 4 Strandburg K. 114 Raasch C. 55, 81 Striukova L. 123 Rademaker D. 147 Su Y. 148 Randhawa K. 35, 36 Tekic A. 112 Rau C. 39 Tekic Z. 40 Rayna T. 123 Thøgersen J. 155 Redlich T. 76 Tietze F. 189 Reetz D. 92 Tokic S. 113 Reimo J. 7, 132 Topic P. 52 Reinsberger K. 15, 101, 163 Van Alstyne M. 187 Reith I. 52 Van Dycke L. 115 Rindfleisch A. 94 Van Oers L. M. 9 Roberts D. 45 Van Overwalle G. 115 Rützler L. 28 Voigt K. 88 Sachdev V. 94 Von Hippel C. 167 Sachsendorfer W. 163 Von Hippel E. 195 Saldien 168 Vroljik A. 73 Salge T. O. 34 Wagner T. F. 88 Salter A. 30 Walcher D. 181 Sanchez S. J. 145 Wang N. 97 Schaarschmidt M. 20, 22, 158, West J. 36, Schewe G. 60 Wilden R. 35, 36 Schlottmann M. 174 Willoughby K. 112 Schmidthuber L. 121 Wimbauer L. K. 70 Schöttl C. P 81 Winding J. 41 Schrader M. 170 Witzel A. 151 Schultz C. 189 Wu C. 55 Schweinsfurth T. 55 Wulfsberg J. 76 Schwemmann B. 68 Young A. 170 Seeber I. 29 Zaggl M. A. 81 Seymor A. 183 Zantedeschi D. 29 Shrestha Y. R. 12 Zejnilovic L. 171 Silva J. 169 Zeng M. A. 7, 132 Smolka C. 188 Søndergaard H. A. 155 Sornette D. 2 Spedale S. 45 Spruit K. 198 Stanko M. 146 Stock R. 131,159, 160

217