EMERGING BUSINESS MODELS OF THE MOBILE INTERNET MARKET

Transcription

1 HELSINKI UNIVERSITY OF TECHNOLOGY Faculty of Information and Natural Sciences Department of Computer Science and Engineering Zhao Hanbo EMERGING BUSINESS MODELS OF THE MOBILE INTERNET MARKET Master s Thesis Espoo, June 30, 2008 Supervisors: Professor Antti Ylä-Jääski, Helsinki University of Technology Associate Professor Markus Hidell, Royal Institute of Technology Instructor: Professor Heikki Hämmäinen, Helsinki University of Technology Hannu Verkasalo, Lic.Sc. (Tech), Helsinki University of Technology i

2 HELSINKI UNIVERSITY OF TECHNOLOGY Faculty of Information and Natural Sciences Degree Programme of Security and Mobile Computing ABSTRACT OF MASTER S THESIS Author: Zhao Hanbo Date: 30 th June 2008 Pages: 103 Title of Thesis: Emerging business models of the mobile Internet market Professorship: Data Communications Software Code: T-110 Supervisor: Instructor: Professor Antti Ylä-Jääski Associate Professor Markus Hidell Professor Heikki Hämmäinen, Hannu Verkasalo The convergence of the telecommunication industry and the Internet industry results in a transformation of the established value network and the business landscape, creating both opportunities and threats for incumbent players and new entrants. Major uncertainties affecting the telecommunication industry are not technical factors, but rather stem from the evolution of underlying business models. The aim of this thesis is to figure out the business logic behind successful mobile Internet services. Business model is such a broad concept that covers a large rage of aspects. Besides, none of the existing business model analysis framework is universally accepted in the academic world. The present thesis will focus on revenue model study and value network analysis. In this thesis, an analysis framework is developed and applied to the study of mobile Internet services. The framework categorizes all the value exchange into three groups, namely services & goods, monetary benefits and intangible benefits. In addition, a revenue model classification is adopted, in which the revenue models of mobile services are sort out into advertising model, subscription model and transaction model. The study picks out the top twenty mobile Internet services present in the Finnish market on the basis of a set of pre-defined criteria. The results of analysis indicate that generally services adopting advertising model outnumber their counterparts in several aspects. Moreover, a cross-case study among all the advertising model services is conducted, in which the advertising models are further broken down into four variants. They are two-sided market model, double two-sided market model, three-sided market model and hybrid model. Another lesson learned from this study is that the orchestrator of value network varies from case to case, which doesn t necessarily have to be the telecom operator like in the traditional value networks. Keywords: mobile internet, revenue model, value network, two-sided market ii

3 Acknowledgements This study was carried out at the Department of Electrical and Communications Engineering, Helsinki University of Technology. As part of the MoMI project, the thesis was produced in collaboration with other researchers and industrial partners. The study process was challenging but rewarding, in which I learned a lot. First of all, I would like to thank Prof. Heikki Hämmäinen (TKK), Prof. Antti Ylä-Jääski (TKK) and Associate Prof. Markus Hidell (KTH). Their high-level guidance and supervision pointed out a clear direction for my research. Also, I would like to express my gratitude to my instructor, Hannu Verkasalo who had been working closely with me. His instructions provided a great deal of help for my study and always kept me on the right track. In addition, the support from one colleague of mine, Mikko Heikkinen, is highly appreciated as well. Despite that there was no direct link between his work and mine, he always spent time helping me and suggested a lot of valuable insights. Furthermore, I would also like to thank all my colleagues in the laboratory, Antero Kivi, Timo Smura, Niklas Tirkkonen, Borja Jimenez Salmeron, Turo Brunou and Matti Uronen for their support of my work. Finally, I would like to address my deepest gratitude to my family and friends for providing support during my studies. Espoo, June 2008 Zhao Hanbo iii

4 Abbreviations and Acronyms 2G Second Generation 3G Third Generation API Application Programming Interfaces ARPU Average Revenue Per User DNA Deoxyribonucleic Acid GPRS General Packet Radio Service GPS Global Positioning System GSM Global System for Mobile communications ICT Information and Communication Technology IP Internet Protocol ISP Internet Service Provider LAN Local Area Network MMS Multimedia Messaging Service MNO Mobile Network Operator MoMI Modeling of Mobile Internet Usage and Business MSN Microsoft Network MVNO Mobile Virtual Network Operator PC Personal Computer PDA Personal Digital Assistant PDF Portable Document Format SIP Session Initiation Protocol SMS Short Message Service TV Television VoIP Voice over Internet Protocol WiMAX Worldwide interoperability of Microwave Access WLAN Wireless Local Area Network iv

5 Contents ACKNOWLEDGEMENTS... III ABBREVIATIONS AND ACRONYMS... IV CHAPTER1 INTRODUCTION BACKGROUND PROBLEM DEFINITION OBJECTIVES AND SCOPE METHODS STRUCTURE...6 CHAPTER 2 THEORETICAL OVERVIEW VALUE NETWORK REVENUE MODEL TWO-SIDED MARKET ATTENTION ECONOMICS RESOURCE-BASED VIEW OF THE FIRM TRANSACTION COST ECONOMICS DOUBLE HELIX MODEL...23 CHAPTER 3 INDUSTRIAL BACKGROUND TELECOM INDUSTRY STRUCTURE Increasing players and competition Evolving industrial structure Comparison of regional markets REVENUE MODELS OF MOBILE AND INTERNET SERVICES Mobile services revenue models Revenue models of Internet services Comparison of mobile and Internet services...35 CHAPTER4 ANALYSIS OF MOBILE INTERNET SERVICES ANALYSIS FRAMEWORK Case selection Within-case study Cross-case study Value network model DESCRIPTION OF SELECTED SERVICES AND ACTORS Selected services Industrial actors...51 v

6 4.3 DESCRIPTIVE ANALYSIS Comparison of revenue models Comparison of company market shares VALUE NETWORK STUDY Visual Radio Internet Radio Nokia Maps Nokia Search Opera Mini MSN Google Maps Lifeblog Navicore, Nokia Navigation Opera Mobile Nokia Channels Google Mail Nokia WidSets Elisa Mobi TomTom Nokia Photo Print Nokia Music store CLASSIFICATION OF ADVERTISING MODELS Two-sided market Double two-sided market Three-sided market Hybrid...74 CHAPTER 5 CONCLUSION SUMMARY AND DISCUSSION LIMITATIONS FUTURE RESEARCH...81 BIBLIOGRAPHY...83 APPENDIX A...90 vi

7 List of Tables TABLE 4.1 SELECTED SERVICES...40 TABLE 4.2 COMPARISON BY CATEGORIES OF REVENUE MODEL...53 vii

8 List of Figures FIGURE 3.1 THE NEW TELECOMMUNICATION INDUSTRY...25 FIGURE 3.2 3G VALUE CHAIN...26 FIGURE 3.3 THE INTERWOVEN VALUE NETWORK AND VALUE CHAIN...27 FIGURE 4.1 VALUE EXCHANGE OF GOOGLE SEARCH...44 FIGURE 4.2 COMPARISON OF MEAN CUSTOMER CONTENTION RATES...54 FIGURE 4.3 COMPARISON OF NUMBER OF TRYERS AND USERS BY COMPANY...55 FIGURE 4.4 VALUE NETWORK OF VISUAL RADIO...56 FIGURE 4.5 VALUE NETWORK OF INTERNET RADIO...58 FIGURE 4.6 VALUE NETWORK A OF NOKIA MAPS...59 FIGURE 4.7 VALUE NETWORK B OF NOKIA MAPS...60 FIGURE 4.8 VALUE NETWORK C OF NOKIA MAPS...61 FIGURE 4.9 VALUE NETWORK OF NOKIA SEARCH...61 FIGURE 4.10 VALUE NETWORK A OF OPERA MINI...63 FIGURE 4.11 VALUE NETWORK B OF OPERA MINI...63 FIGURE 4.12 VALUE NETWORK OF MSN...64 FIGURE 4.13 VALUE NETWORK OF GOOGLE MAPS...65 FIGURE 4.14 TWO-SIDED MARKET VARIANT A...70 FIGURE 4.15 TWO-SIDED MARKET VARIANT B...70 FIGURE 4.16 TWO-SIDED MARKET VARIANT C...71 FIGURE 4.17 DOUBLE TWO-SIDED MARKET...72 FIGURE 4.18 THREE-SIDED MARKET...73 FIGURE 4.19 EXAMPLE OF HYBRID MODEL...74 viii

9 Chapter 1 Introduction In the beginning of this chapter, the background of this research is presented. Then several key research questions as well as the objectives and scope of this thesis are described. After that are the methods adopted, by which the preceding research questions are given answers to. Finally, the structure of this thesis is outlined. 1.1 Background The continuous evolution of information technology has been changing the way people lead their lives over the last century. Mobile communication technology and the Internet are probably among the most influential technologies. As both industries have experienced an upturn, it is widely believed that convergence of mobile communication and the Internet, namely the mobile Internet, will gain extensive popularity as its predecessors did. For the telecommunication industry, prevalence of mobile internet might turn into a solution to its declining ARPU (Average revenue per user) and promises an end to commoditization of mobile services [60]. For the Internet industry, mobile Internet technology eliminates the spatial and temporal limitation of Internet and enables users a broader access. The cellular subscriber has been outnumbering Internet users worldwide for almost two decades and increasing more rapidly [25]. It was predicted by Nokia, the world leading mobile phone manufacturer, that the total number of mobile subscribers would outstrip three billion in the year of 2008 while there were only one billion Internet subscribers at the time the prediction was made [38]. Some experts advocate many people would probably get their first taste of Internet through mobile phones in developing countries 1

10 CHAPTER 1. INTRODUCTION 2 where mobile access is more economically feasible and affordable for local residents [58]. As the legacy business models become rather mature, prospects of margin and profit are very limited. Thus, incumbent actors in telecommunication industry are making attempts of new business models. On the other hand, a bunch of new entrants are keen to challenge the market, and some of them are prepared with ground-breaking business models. Recent actions by BLYK, iphone and Nokia Ovi exemplify innovation of business model. Most of the successful value added services today are SMS-based. However, there is tremendous potential of value creation and profit generation through innovative services that take advantage of the technology technical evolution. This research is a part of MoMI project (Modeling of Mobile Internet Usage and Business), which is aimed at a better understanding of the evolution of telecommunication industry and mobile service market from a techno-economic perspective [35]. Not only do the participants of MoMI project represent a comprehensive coverage of actors in Finnish telecommunication market, including handset manufacturer (Nokia), mobile operators (Elisa, DNA, Ainacom), regulation authority (Finnish Ministry of Transport and Communications), consulting firms (Accenture) and broadcasting company (YLE), but they also include foreign research institutions from e.g. U.S. and Japan. The project collects empirical data of Finnish mobile market and adopts pioneering data mining methods to analyze them. On the basis of the data analysis, both quantitative and qualitative studies are conducted to explain the evolution of mobile business models.

11 CHAPTER 1. INTRODUCTION Problem definition To date, a great deal of mobile Internet services has been introduced in different places around the world. Some of them enjoy excellent performance in terms of profitability while the others struggle to survive or fail to attract adequate users to cover the cost. It is not yet clear that what are the common characteristics of the winners and that of the underdogs. Up until this point, many would agree that the only success of mobile internet around the world is NTT DoCoMo s i-mode. In contrast, the outcomes are well below the initial expectations in Europe and the U.S. It reaches broad consensus that there is no universal business model that can apply to all the cases arising from the various factors, for instance economical, social, demographical and political factors, in different regions. As mentioned above, there is no clear picture yet how future business model in telecommunication industry will look like. Many actors have been looking for the answer, by pioneering new ways of doing their business. The situation gives rise to the key question that this research will study: What kind of business model makes some mobile Internet services more successful than the others and how do these business models change the business environment? The concept of business model itself, however, doesn t have a commonly accepted definition. The generic schools of thoughts of business models have limitations when applied to cutting edge business scenarios. Most of the existing studies are for describing business opportunities rather than an academic research method. The elements of a business model consist of value proposition, market segment, value chain within the firm, cost structure and profit potential, value network as well as competitive strategy [9]. In this research, the focus will be revenue modeling and high-level value

12 CHAPTER 1. INTRODUCTION 4 network modeling instead of the other aspects. Hence the main research question can be broken down into following sub-questions: 1. How do revenue models affect the performance of mobile Internet services? 2. What kind of impact do relationships among different actors of a value network have on the performance of a mobile Internet service? 3. How the business models of different mobile Internet services differ from each other? 1.3 Objectives and scope The objectives of this thesis were defined in order to answer the research questions mentioned above. The blueprint of this analysis can be decomposed into four milestones. 1. Define criteria of evaluating performance of mobile Internet services and pick out research objects (i.e. services) 2. Conduct revenue model analysis and high-level value network analysis of each selected service 3. Generalize the findings, find common characteristics and differences of those business models, and then categorize them into groups 4. Identify the development trend of mobile Internet services on the basis of obtained results Before unfolding this research of mobile Internet, it is crucial to clarify what the term mobile Internet exactly stands for. In a broad sense, mobile Internet could refer to any packet switched data transmission based on IP (Internet protocol) protocol via any kind of portable computing device (incl. laptop). Nevertheless, the definition is narrowed down in this study. By mobile, we mean portable handheld computing devices with

13 CHAPTER 1. INTRODUCTION 5 cellular communication capability, which consist of regular mobile phones, smartphones, some of the PDAs (Personal Digital Assistant) and etc. By Internet, on the other hand, we indicate communications and data transmissions on the basis of IP protocol suite through either cellular network or WLAN (Wireless Local Area Network). Moreover, the mobile Internet services in this research further concentrate on services that can be accessed at least via cellular data transmission network. They can be offered by either mobile operators or third-party service providers. Those non-data services (e.g. voice call, SMS) and standalone applications (e.g. alarm clock) are out of the scope of this research in spite of their popularity. As part of MoMI project, this study leverages the empirical data of Finnish mobile market, but only for the purpose of service evaluation and service selection. Those mobile Internet services are evaluated judging from their performances in Finnish market. Thus distinct features of Finnish market could influence the selection of services. The analysis of business model, on the other hand, bases on a global scope. Though serving Finnish customers, the value networks of those services don t necessarily have to be made up of local firms. In other words, the appearance of foreign companies is inevitable. All in all, revenue models and value network formulation will be investigated from a broader perspective than the selection process. 1.4 Methods Five main research methods are applied in this thesis. These are: Literature review Expert interviews Value network analysis framework

14 CHAPTER 1. INTRODUCTION 6 Empirical data analysis and case studies Analysis based on the theoretical framework and case study results Literature review is intended to build up basic background for this thesis work. Above all, it provides a solid foundation for the following analysis that a reasonable understanding of different theoretical frameworks as well as the state-of-art of mobile Internet industry. In addition to the literature study, expert interviews (mainly the researchers working in MoMI) are conducted to get a better understanding of theoretical frameworks and complicated business landscape of mobile Internet industry. In the final stage, the very experts are asked for comments on the studies done so as to improve the research work. After acquiring fundamental knowledge, the research carries on with selection of mobile Internet services to be analyzed in the following step. Drawing on empirical data analysis, we present usage of all the services and applications in the dataset from multiple dimensions. Then the criteria of selection are defined, and as a result a set of services is chosen for further case study. Finally, a generalization is built upon both the theoretical framework and findings from case studies. 1.5 Structure The big picture of this thesis is like follows: after chapter1, the introductory chapter, chapter 2 and chapter 3 are two literature review chapters which introduce the theoretical framework and industrial background respectively. In chapter 4, in which most of the unique contribution of this thesis is presented, we explicate the results of our analysis. In the final chapter, conclusions of the study are drawn together and the potential future research is discussed.

15 CHAPTER 1. INTRODUCTION 7 Chapter 2 Theoretical overview This chapter presents the theoretical cornerstone used for the following analysis. It is divided into seven subsections. In the first two parts, value network and revenue model, which constitute the core theoretical foundation of this study, are elaborated. Then we will move on to the theories of two-sided market and attention economics in the following two subchapters. They will be applied with the intention of analyzing and explaining some of the advertisement-oriented business models. After that, this chapter will carry on with resource based view of the firm and transaction cost economy, both of which provide insightful suggestions for explaining strategic network formulation. The last subsection is about double helix model that was pioneered to illuminate dynamics and evolution of telecommunication industry. Chapter 3 Industrial background The focus of this chapter will be on telecommunication industry rather than the Internet industry. After all, mobile Internet primarily evolves out of telecommunication industry. First, industrial landscape of telecommunication world is portrayed, including state-of-the-art, diffusion of technologies and different actors involved in the show. Second, the existing business models of both telecommunication and Internet industries are illustrated. Third, a comparison of two industries is made to give implication how to apply research findings of Internet industry to telecommunication industry. Chapter 4 Analysis of mobile Internet services

16 CHAPTER 1. INTRODUCTION 8 This chapter starts with a brief introduction to the analysis procedures of our research. Following is the selection of services and description of the services chosen to be analyzed As for the selection process, an assessment of those services is explained with support of empirical data analysis. Thereafter all the selected services are studies case by case. Finally, the findings of case studies are generalized and a comprehensive summary is presented. Chapter 5 Conclusion In this chapter a conclusion is drawn to answer the questions defined in the introduction chapter and summarize the whole research and the analysis. Besides, we also discuss the limitations and suggest possible directions for future research.

17 Chapter 2 Theoretical overview Before diving into the analysis of mobile Internet industry, various sources of business model analysis from a range of theoretical perspectives are gone through. The literature review covers value network, revenue model, two-sided market, attention economics, resource based view of the firm, transaction cost economy, as well as double helix model. Each one of those perspectives is elaborated and then argued that how it relates to this thesis. 2.1 Value network Introduced by Porter in mid-1980s [44], the concept has evolved from the linear value chain at the beginning to the multi-dimensional value network. There are a number of buzzwords around these days which are inherently similar to value network, including value constellation, strategic network, business ecosystem, cluster, etc. In Porter s value chain analysis [44], the production of companies is identified as primary activities and support activities over the value chain. The mission of primary activities is to deliver the value proposition, which exceeds the cost and hence bring about margin, to the end customer. Primary activities incorporate inbound logistics, operation, outbound logistics, marketing and sales, and services. They directly affect value creation of a company while support activities have indirect influence, only by affecting primary activities. Support activities consist of infrastructure of the company, human resource management, technology development and procurement. The performance of both kinds of activities determines the fate of a company. Competitive advantage can be achieved either by the capability of having a lower cost or better 9

18 CHAPTER 2. THEORETICAL OVERVIEW 10 differentiation than its rivals. In the Report on Value Chains [48], four key factors are identified as determinants for successful participation or integration of enterprises in existing value chains. They are entry barriers, rents, governance and upgrading. Barriers to entry and rents are asserted that the winning strategy is to develop inimitable competitive advantage and exploit it to generate economic rents. They argue that the actors who capture the majority of the value are the ones that are capable of protecting themselves from competition by possessing scarce attributes and involving barriers to entry. The concept value chain governance indicates the bargaining power and influence one cooperates owns over the other actors on the value chain. It is a reasonable explanation for who has the market access, why some firms have to rapidly improve their production capabilities, the way gains are distributed along the value chain and the impact of value chains over policy makers. The concept of upgrading stresses that the speed of innovation is critical to win over the competitors. This development of value chain analysis can be useful in business analysis of mobile services. For instance, Nokia decided to put a great effort on developing services and bundle those services with their handsets. By doing so, Nokia may increase its governance on the value chain and raise the entry barrier of both handset manufacture and service provision. The approach of value chain analysis has been questioned if it could apply to all the sectors of economy. Normann and Ramirez [39] raised the value chain analysis was grounded upon the hypothesis and model of traditional industries where the value chain is already mature and stable. As an alternative, they proposed the concept of value constellation, within which different economic actors suppliers, business partners, allies, customers work together to co-produce value. The strategic decision making should not be confined to a single firm level, they stress, and it rather ought to be the

19 CHAPTER 2. THEORETICAL OVERVIEW 11 reconfiguration of roles and relationships among this constellation of actors in order to mobilize the creation of value in new forms and by new players. As a synonym of value constellation, value network is defined by Timmers [53] that value network is a multi-enterprise network of relationships focused on integration of information flows to exploit information and knowledge in the network for strategic business objectives. Likewise, strategic network is termed as stable inter-organizational ties which are strategically important to participating firms. They may take the form of strategic alliances, joint ventures, long-term buyer-supplier partnerships, and other ties. [21] In his book Mobile services in the networked economy, Vesa [56] points out those theories are helpful in the strategy research of mobile industry because it involves a range of social actors and the business system is constantly changing. Gulati et al. [21] further elaborate the major benefits a firm can gain by becoming a part of a strategic network. First of all, the tight inter-organizational relation decrease competition and raise the barrier to entry. Besides, the firms which take vantage points called structural holes enjoy bigger profitability. Secondly, along with barrier to entry, strategic alliance also sets up mobility barrier for participating members which prevent them from switching to other alliances. Thirdly, the uniqueness and inimitability of network one company is in may contribute to its comparative advantage. Fourthly, network can take edge off the transaction cost within the network by decreasing the information asymmetry and increasing the cost of opportunism. Lastly, the stability and dynamics vary from one network to another. Learning race can be a good illustration where an actor quit the network and leverage on its own competence once it has learned all the capabilities its partners have. Cluster represents a phenomenon of geographical concentration and proximity of linked companies in a certain industry. [45] Two factors make up the superiority of

20 CHAPTER 2. THEORETICAL OVERVIEW 12 cluster. One is spillover that firms within a cluster benefit from the flow of information and diffusion of innovation. The other is the fierce competition within a cluster encourage and challenge all the actors there to perform at their highest levels. In addition, Porter points out the evolution of clusters or the emergence of new clusters can derive from one or two leading companies which stimulate the development of the whole cluster. It can be exemplified by the Japanese operator driven clusters, Finnish Nokia driven cluster, Swedish Ericsson driven cluster and so forth. Business ecosystem is another interesting concept of strategic research. According to Moore [36], business ecosystem is stated as an economic community supported by foundation of interacting organizations and individuals. Unlike the concept of cluster, business ecosystem doesn t emphasize the co-location of actors since the development of information technology and globalization reduces the importance of spatial proximity. On the other hand, Moore introduces the life cycle of business ecosystem which consists of birth, expansion, leadership and self-renewal or death. Each one of those four phases is illustrated with cases from mobile industry in Jarkko Vesa s book [56]. The preceding concepts have a lot in common whereas each of them has some distinct characteristics. In the comparison from Peltoniemi [42], two differences are noteworthy. The first one is about the different views on competition and cooperation in those models. Clusters gain power from severe internal rivalry and value networks focus on cooperation while business ecosystems involve both competition and cooperation. The second dissimilarity is knowledge sharing. Knowledge flows within clusters because all the actors monitor each other and make quick response to others changes. On contrary, members of value networks are somehow willing to co-operatively create or transfer knowledge but only to a limited level. In business ecosystem, knowledge sharing and

21 CHAPTER 2. THEORETICAL OVERVIEW 13 co-production are enabled by the interconnectedness and motivated by the shared fate. 2.2 Revenue model Different schools of thoughts have a variety of definition of business models and revenue models, so as the scope of them. Some scholars advocate that business model and revenue model are two distinct approaches for business analysis while the majority believes revenue model is an element of business model. According to Amit and Zott [3], the main concern of business model is value creation whereas the heart of revenue model is value appropriation. They claim that the concept of business model and revenue model are complementary but different. Nevertheless more specialists consider revenue model as a constituent part of the business model. Osterwalder and Pigneur [40] define revenue model as the capability of translating customer value proposition into cash flow, namely the incoming revenue stream. Combined with cost structure and profit model, revenue model is classified into the financial aspect of their business model framework. Petrovic et al. [43] divide a business system into seven sub-models, not least of which is revenue model. They describe revenue model as the logic of what, when, why and how firms gain compensation for their products. Likewise, Mahadevan [30] decompose business model into three streams, i.e. value stream, revenue stream and logistics stream. Focusing on the economic and financial aspect, Kim and Marbourgne [27] highlight price and revenue models are essential for ensuring profitability. In this thesis, we consult the latter perspective that business model goes beyond revenue model and embraces it as a constituent. A comprehensive framework of business model also covers other aspects like target customer, partnership network, etc. Unlike the established theories, this thesis however analyzes revenue models of

22 CHAPTER 2. THEORETICAL OVERVIEW 14 different value networks, of which various actors work together to provide services, in addition to revenue models of each individual firm. The revenue model analysis is conducted from two different levels. One level is how a revenue model is interpreted by users, namely the pricing policy of one service. It serves as the interface between the value network and end-users. The value network here is regarded as a black box, of which the outputs are products (services) and pricing policy while the inputs are money and attention (see chapter 2.4 attention economics) paid by users. On the second level, on the other hand, the internal structure of the black box is analyzed. Generally, it is about the relationship between different value network members, or rather, how those inputs are shared among them. To sum up, the first level is to expound value appraisal and value appropriation whereas the second level is to elaborate value allocation and value generation. Among all concepts of revenue models, a classification from Amit and Zott [3] is adopted in our analysis. According to them, the revenue generation of Internet services can be categorized into three groups. They are subscription model (S), advertising model (A) and transactional model (T). The transaction model is comprised of fixed transaction fees, referral fees, fixed or variable sales of goods, etc. Various variants of these three models can be seen. Besides they are not exclusive so that they can be used in a combination as well. This classification is chosen due to its conciseness and comprehensiveness. It covers revenue models of Internet services, which are inherently similar to mobile Internet services, without bringing about further confusion. In this research, we will only carry out qualitative analysis from a strategic level. Details of revenue models and trivial differences are not of our greatest interest. Therefore this SAT model meets exactly what we need.

23 CHAPTER 2. THEORETICAL OVERVIEW Two-sided market Several papers yield insightful suggestion for this research regarding two-sided market (e.g. [15] [16] [47]). Two-sided markets are defined as markets in which there are two distinct sets of customers, each of which exerts a positive network externality on the other group over a common platform. Platforms, in this scenario, are the services and products which gather groups of users in two-sided networks [15]. In a two-sided market, customers of each group exhibit a preference to the platform which has more users in the other group. A number of industries can be regarded as two-sided market. Credit card issuers such as Visa and MasterCard need to gain support from massive merchants in order to convince customers to adopt their services, while merchants prefer cards with a wide customer base. Operating systems court developers to build more software running on top of them, and on the other hand vying for more end-users to give those developers enough incentive. Advertising-supported media like TV channels, web portals and newspapers contends for the attention of both viewers and advertisers. Clubs, which provide a platform for men and women to interact and potentially date, need to bring together visitors of each gender. The same virtuous cycle transpires in most successful cases, more demand from one side stimulates the need of the other side [15]. Let s take Sony s video game console PlayStation as an example. The more game developers (one side) work on games for PlayStation platform, the more customers (the other side) will favor this platform thanks to a higher quality and a greater variety of games. In the meantime, a bigger customer base of PlayStation console could encourage developers to join the platform because they foresee increasing sales of their games. The key factor of two-sided market, according to Eisenmann et al, is to get pricing right [15]. Unlike traditional pricing approaches such as cost-based pricing, value-based

24 CHAPTER 2. THEORETICAL OVERVIEW 16 pricing or competition-based pricing, pricing in two-sided platforms is in need of more complicated consideration. Platform providers need to be cautious in pricing of both sides. They may have significant impacts on each other and therefore cannot be considered separately. A proper pricing structure must be deliberately designed in order to get both sides on board [47]. Normally platform providers reduce the prices for the subsidy side, of which the increasing size makes the money side, i.e. the other user group, willing to pay a premium to access this platform [15]. The prices for the subsidy side are usually set lower than the level if it were regarded as independent market. The presumption is a large size of subsidy side provides a solid foundation to yield network effect. On the other hand, the money side is charged more than they would be in an independent market to make up for the loss from the subsidy side, that is to say the platform provider earns more on the whole. The major challenge is to precisely determine the prices of each side in order to maximize the overall profit. To find the most optimal pricing policy, several dimensions should be taken into account for decision-making of platform providers [15]: Ability to capture cross-side network effect: The platform providers need to make sure that they have fully control on the subsidy side they are spending money on [15]. The platform should be designed to be incompatible with other money side providers. Otherwise, subsidy side users have the opportunity to take advantage of the giveaway while choosing other money side providers. A negative example is Netscape. Hoping to sell its servers to websites, Netscape gave its browser away for free to the end users, which are considered as subsidy side in this scenario. Nevertheless, web sites could use servers from other providers to generate the same web pages, and therefore the strategy fell to the ground.

25 CHAPTER 2. THEORETICAL OVERVIEW 17 User sensitivity to price: Typically the user group which is more sensitive to price is treated as subsidy side [15]. A relatively small amount of subsidization might give a great rise to the customer base which makes it a more cost-effective approach. Typically advertisement-oriented dot-com companies offer free services because users get used to free lunch on the Internet. They may switch to other service providers even if only a tiny amount of fee is charged. Output cost: The user group with zero or near-zero marginal cost is usually subsidized [15]. It is especially common in digital content provision. Portals like Yahoo supply free access to their content so as to attract more eyeballs. Nonetheless, platform providers need to be cautious with decision-making in the case the marginal cost is not neglectable. A negative case given by Eisenmann et al. is FreePC [15]. It offered free PCs bundled with Internet connection in purpose of getting giveaway-takers as advertisement audience. Yet those low-profile customers are not of marketers high interest. Consequently FreePC didn t manage to cover the expensive cost. Same-side network effects: Users of each side of the platform probably have preferences in terms of the size of the group they belong to. Same-side network effects can be positive as snowballing [15]. For instance, the more users purchase Xbox console, the easier players can find partners to play online games or players to trade games with. Should there be competition between members from the same group, negative same-side network effect may then come into being. If it is particularly strong on one side, it is sensible to charge that side or limit the number of players. E.g. Autobytel gives exclusive access to one dealer in each region and charges clearly for that. User brand value: Getting some celebrity users on board could greatly increase

26 CHAPTER 2. THEORETICAL OVERVIEW 18 attractiveness of the platform [15]. Those celebrity users can be the ones with big volume like big buyers and suppliers. Let s take digital music business as an example. With the aim of making its Music Store successful, Nokia has been trying hard to persuade big labels to join its platform. Not only does the involvement of big record companies make supply of music swell which is appealing to end users, but it also serves as a good reference so that it is easier to convince smaller labels to get onboard. The theory of two-sided market contributes valuable insights for analysis of mobile Internet services. It has been applied to studies of both telecommunication industry and Internet industry [10] [55] [61]. The conventional economic school of thoughts, in a lot of cases, cannot explain dynamics between operating systems, content providers, advertisers, handset manufacturers and network operators very effectively. Especially when it comes to the case of pricing policy, two-sided market theory can properly account for the free or near-free service provision. The weakness of this theory is, however, no common theoretical framework has been widely accepted so far. 2.4 Attention economics It is universally acknowledged that we live in an era of information explosion. Our eyes and ears all always flooded with different sources of information. As a consequence of overloaded information, most of us have experienced an attention crisis, that is to say we don t know where to spend our attention on. News that used to take several minutes to read through now only gets a few seconds, because people just skim over it. The phenomenon is precisely explained by Simon [52] that the explosion of information results in the scarcity of attention. In many consumer-oriented industries, attention is turning into the most scarce resource competitors contending for. Essentially it is reshaping value chains, or value networks, of many sectors it is involved.

27 CHAPTER 2. THEORETICAL OVERVIEW 19 Let s take media industry, which is perhaps the most prominent in terms of attention economy, as an example [51]. Stepping back to the mass media world, distribution was the scarcest resource across the value chain. The scarcity in some sections is resulted from regulation. The usage of distribution channel (e.g. radio, TV broadcast) must be approved by government authorities. In other segments, natural monopoly gives rise to the scarcity (e.g. newspaper). As time went on, new technologies had been restructuring those sectors, by changing attention into the scarcest resource. Those technologies enable a number of novel ways of production, which can be exemplified by various user-generated contents like blog, and distribution, of which the most remarkable example is undoubtedly the Internet. In other words, production and distribution experience more abundance than attention. Consequently the competition for attention is getting fierce: attention is becoming more valuable than production and distribution. In attention economy, customers accept free or price-reduced services in exchange of their attention [18]. The most significant difference of attention economy from traditional ones is there is no direct monetary transaction involved in the attention marketplace, namely between service providers and end users. Instead, service providers monetize the attention from customers by selling it to third parties such as ad agency. This is essentially how attention economics are applied in this study. To take a step further, we deem attention as a common currency from a broader sense. Service providers who directly obtain the attention from end users do not necessarily monetize it. As an alternative, they could trade with another actor in the value network to get other types of strategic asset. But eventually, attention must reach one end of the value network, i.e. the ad agencies; no matter it is sold by whom. Three factors, according to Iskold [24], are vital in attention economy. The first one is

28 CHAPTER 2. THEORETICAL OVERVIEW 20 happiness. The abundance of information is annoying because it pushes up the information searching cost. People always need to make a choice where to pay attention to. A successful attention economy brings customers happiness by showing them the very information they are looking for. The second crucial element is relevance. The more relevant one site s content is to visitors expectation, the higher likelihood they will offer their attention to this site. Also, users probably stay there for a longer time or are more likely to click the ads if the content is relevant. Another aspect is privacy which emphasize users need to have control on their personal information in addition to getting protection of it. An important measurement of attention, addresses Davenport and Beck [13], is stickiness. The competition for attention is a zero-sum game. Acquisition of attention of one site means losses of others. A good site is one that capable of not just attracting visitors, but also keeping them coming back and spending more time on the site. A successful site, according to Davenport and Beck [13], must have outstanding performance in four dimensions, relevance, engagement, community and convenience. Relevance indicates that the services must fulfill users needs. It could be either versatile like Yahoo, targeting multiple user groups, or specifically targets at a vertical market. Besides, the needs of users are dynamic which could be different from time to time. The services also need to be adjusted rapidly to adapt to the change. In addition to relevance, a popular site needs get ahead in engaging its users. Several tactics can be leveraged to achieve that such as enhancing interactivity and introducing competition. Another critical issue is community. A strong sense of belonging and ownership prevent users from leaving away. Last but not the least indicator is convenience. An easy-to-use hassle-free service results in great user satisfaction, and thus keeps a high customer retention rate.

29 CHAPTER 2. THEORETICAL OVERVIEW Resource-based view of the firm Firms are regarded as a marshalling and combination of resources and capabilities in the resource-based view [64]. The assumption is different companies possess different resources and capabilities. The products or services, as a result of distinct bundle of services and capabilities, may result in value creation and competitive advantage. Resource-based view was initially established for analysis of individual firms and the intra-organizational resources those firms take possession of. Over the past decade, it has been extended to a network level, specifically applying resource-based view to analysis of strategic alliances [12]. Das and Teng claim that the foundation of alliances is the value-creation potential of firm resources that are pooled together [12]. According to them, the competitive advantage results from effective integration of partner firms valuable resources which one company is unable to provide itself. Similar argument is made by Gulati et al. [21] that resource-based view should be developed to look beyond the boundaries of those individual firms. They further assert a firm s network can be thought of as creating inimitable and non-substitutable value as an inimitable resource by itself and as a means to access inimitable resources and capabilities [21]. If a firm belongs to a network its partners don t, the access to those unique resources of this network, including information, marketing channels, capitals, etc. may make contribution to the firm s competitive advantage. In addition, they point out three types of so-called network resources. The first one is network structure. The basic idea is firms get access to information not only from the actors they directly communicate with but also from the ties of the actors to whom they are connected [21]. The second class is network members. The alliance with a resourceful partner is likely to bring benefit to the focal firm. Last category is tagged tie modality. Various

30 CHAPTER 2. THEORETICAL OVERVIEW 22 characteristics, say strength, closeness and dimension, of the ties one company maintains in its network may all have an effect on its own performance. All the above viewpoints of resource-based view from a network perspective provide thoughtful implication for this thesis. 2.6 Transaction cost economics The main idea of transaction cost theory [66][67][68][69][70] is firms choose between purchasing certain operations from (or outsourcing them to) suppliers and internalizing those activities depending on which one of the two approaches could result in a lower cost, to be exact, pursing the most economically efficient way to conduct operations. Transaction costs incorporate the time and effort spent on searching information, costs of negotiation and establishment of the transaction, costs of production and inventory management, etc. The transaction cost economics well explain the structural evolution from value chain to value network in some industries [60]. The Internet helps reduce the costs of economic transactions. It propels intricate value networks, in which firms dedicate themselves to certain segment and integration of separate activities inside one organization becomes economically suboptimal [60], to replace simple value chains. This kind of progression toward value network is taking place in the present mobile Internet industry. Chan-Olmsted & Jamison [8] claim that there are two approaches of expansion of a telecom firm. One is to develop and provide new services or products independently, only leveraging on its own resources while the alternative is to cooperate and collaborate with other companies. They argue that in the former way a firm can enjoy more powerful control on the network and less dependency on the other players. The consequences are higher cost to bear and greater risk to take. In the latter approach, the

31 CHAPTER 2. THEORETICAL OVERVIEW 23 inducements for companies to join the network are the potential benefits, including access to particular knowledge, sharing risks, strategic synergy and joint venture. A firm evaluates the transaction cost of participating in the network then makes its choice to purse a higher return on investment. 2.7 Double helix model The framework is named after the molecular structure of DNA chains. The idea is the evolution of business world resembles development of natural ecosystem. The evolvement of industries always follows an evolving route of ascendant helix, shifting between horizontal and vertical structure all the time. The initial case studies applying double helix model are mostly from product-related industries whereas Vesa [56] heavily leverages the model for the analysis of mobile industry in his book. According to Vesa, this is because the model seemed to capture nicely the unique characteristics of the Japanese and the Finnish mobile services business model. The oscillation between horizontal and vertical models in mobile industry has also been exemplified by Vesa. He quotes the statement from Fine [17] that firms could often not be content just to stay in one part of the value chain while it is powerful enough to have some influence on the market, but tried to expand vertically and take a piece from others cakes. Then Vesa cites examples like Nokia offers portal and mobile services which used to be provided by operators, Vodafone starts to sell cell phones under its own brand and so forth. On the other hand, a firm cannot benefit from economy of scale and scope once it reaches certain size because the increasing complexity of management adds up to the marginal cost. In the meanwhile, numerous niche competitors are vying to grab a slice of the market from the leading giants. Vesa points out that Nokia has already foreseen one of the major challenges would be managing the growth.

32 CHAPTER 2. THEORETICAL OVERVIEW 24 Double helix model doesn t help in analyzing individual services. It is, however, quite useful for explaining the dynamics and evolution from a long term perspective. In other words, it well depicts the trend of development and changes on a macroscopic level.

33 Chapter 3 Industrial background 3.1 Telecom industry structure Increasing players and competition One prominent feature of the evolution of the telecommunication industry is escalating complexity and competition. To illustrate the intensifying complexity of industrial structure, Fransman [19] argues that it has changed from the previous three layer construction to six layers. The new six layer division (figure 3.1) consists of equipment & software, network, connectivity, navigation & middleware, applications (content is included) and customers. The growth in the number of layers has shaken those closed telecom systems and opens up new opportunities. Not only does it lower the entry barrier for new entrants to get into equipment manufacture or network operator domain, but it also attracts companies from other industries to take their competitive advantages to expand their power to this market. Figure 3.1 The new telecommunication industry (adopted from [29]) Synthesizing value chain of mobile telephony and that of Internet, Maitland et al. [31] draw a value chain of 3G mobile data services (figure 3.2). They highlight a concept of enabler, as a facilitator between hardware manufacture and the services provisioned to 25

34 CHAPTER 3. INDUSTRIAL BACKGROUND 26 end customers, which incorporates middleware, content and application providers. It is pointed out that separation of this layer offers newcomer opportunities to get in on the action and enable the possibility of intermediaries like aggregators. Figure 3.2 3G value chain (adopted from [31]) The traditional structure has collapsed in the telecommunication industry. Several new business models, as a consequence, come into our sights [29]. Device manufacturers may expand into operating system market. Joining together and establishing operating systems and software platforms, handset manufacturers enhance their dominance in the mobile device market and also create potential of new revenue sources from value added services. Software companies extend their influence by conducting investment and acquisitions upon upstream manufacturers. Financial service firms, like credit card issuers and banks, take advantage of their strong customer relationship and provide them customized handsets with capabilities of M-banking and M-commerce. Vertical expansion and integration are not, however, the only option for actors. Having not integrated with any infrastructure or device manufacturer, Yahoo maximizes its accessibility and therefore augments its own value as a portal Evolving industrial structure Not just do the rising number of players increase the competition, it has also been reshaping the industrial structure and triggering more complexity. More choices reduce

35 CHAPTER 3. INDUSTRIAL BACKGROUND 27 the dependency of firms on their suppliers, but on the other hand propel them to improve their competitiveness as a response to the fiercer competition. Also, a bigger variety of potential partners firms could ally with result in possibilities of different kinds of strategic network. According to Li and Whalley [29], the multidimensional structure opens up more entry points and exit points for industry actors and gives more choices to the customers (figure 3.3). After shakeout of upstream industries, they claim, the likelihood of new opportunities is moving downward on the value chain, closer to end customers. Figure 3.3 The interwoven value network and value chain (adopted from [29]) Comparison of regional markets While the telecom industry has experienced deconstruction and is being reshaped [29], diversiform structures appear in different markets [56]. Three typical examples can be identified in Japan, Finland and the United Kingdom. The Japanese market takes on a vertical integrated structure where three oligopolistic operators, NTT DoCoMo, KDDI

36 CHAPTER 3. INDUSTRIAL BACKGROUND 28 and Vodafone K.K., provide one-stop services to the subscribers. In this case, mobile subscription, purchase of handsets and value added service subscription are bundled together for customers. That is to say, customers do not have the freedom to choose their handsets, add or cancel any of the value added services because both handsets and services are technologically customized to a specific operator s network and incompatible with other networks. On the contrast, the telecom industry of Finland has presented a horizontal, modular structure. Players don t set foot in other parts of the value chain and concentrate on winning the competition over their counterparts. Part of the reason is thanks to the open unified standard. In other words, a customer can select any mobile subscription, insert the SIM card into any handset on the market, and then choose any value added service without compatibility restrictions. Like neither of the preceding models, the structure of the UK telecom industry appears to be a hybrid of the vertical and horizontal models. Customers can find mobile subscription with neither handset nor additional service whereas there are bundling packages which include specific mobile devices and services as well. In addition to qualitative analysis of the overall structure of the telecom industry in those three markets, Vesa [56] conducts measurement of their value networks. The quality of different networks is evaluated judging from so called strength of the network, which is made up of four elements: the standard alone product performance, the user network, the complements network and the producer network. He argues that the mobile networks exhibit better performance and the handsets there present higher cost-effectiveness for the end consumers. From the content and service perspective, the Japanese providers are undoubtedly much more ahead of their European counterparts. About the user network, the penetration of mobile data service is also higher than in the UK and Finland, not to mention that the population of those European countries cannot compare with Japan. As to the complements network and producer network, Japan

37 CHAPTER 3. INDUSTRIAL BACKGROUND 29 scores higher than the UK and Finland as well. The mobile market attracts complementing partners from various industries, from restaurants to printing kiosks, to get in on the action. From the producer viewpoint, a much bigger number of independent content and service providers is the most prominent feature of Japanese market compared with the UK and Finland. For both two networks, the tight cooperation and close integration ensure the stability of those networks in Japan. As the conclusion, Vesa states that the collaboration networks of Japan are the most powerful one among those three, thanks to the vertical integrated industrial structure there [56]. 3.2 Revenue models of mobile and Internet services Mobile services revenue models According to Coursaris and Hassanein [11], the revenue models of mobile services can be categorized into two broad types, namely customer initiated and non-customer initiated, which altogether contain eight models. The major difference between the two groups is the involvement of end consumer payment, that is to say, if there is any cash flow from customers to any one or more of actors in the value network. Hence, the two categories are separately labeled as customer initiated models and non-customer initiated models. Customer initiated revenue models comprise access, subscription and pay-per-use while non-customer initiated models incorporate advertising, transaction, payment clearing, hosting, and point-of-traffic. All the eight models are described below: Access model [11]: Customers pay the bill from mobile operators in order to access the wireless network. Four different pricing schemes are listed, namely flat rate,

38 CHAPTER 3. INDUSTRIAL BACKGROUND 30 time based, volume based as well as an innovative free access model. Subscription model [11]: Customers purchase mobile value added services in a subscription manner. Pay-per-use [11]: Unlike the subscription model, customers just pay for parts of services that they are interested in to avoid a long-term commitment. For instance, an individual who is not crazy about changing ringtones all the time probably prefer to download a ringtone when he needs instead of subscription of a monthly download. Advertising [11]: Content providers or value-added service providers sell advertising spaces for funding the content or service development. Transaction [11]: Transactions take place while they are flowing on the content supply chain, such as the purchase of content from raw content providers by content aggregators. Payment clearing [11]: It refers to commissions the billing and charging provider charge. Hosting [11]: Content providers outsource hosting of the content if they are lack of this capability. Point-of-traffic [11]: Mobile operators subsidize value-added service providers on the basis of the amount of traffic their services generate.

39 CHAPTER 3. INDUSTRIAL BACKGROUND 31 In the above categorization, revenue models are studies mainly from a single firm perspective. In other words, each one of the value network members is considered separately regarding revenue generation. But still, valuable lessons can be drawn from the work of Coursaris and Hassanein [11]. Plus advertising model, customer initiated models can be regarded as different means of value appropriation for the value network though customers pay attentions as the price levied instead of money in advertising model. The other class, non-customer initiated models make useful suggestions about value creation and value allocation in the network. Another study by Camponovo and Pigneur [5] [6] examines revenue models of different actor groups. Mobile device manufacturers do not only sell devices but also make money by operating portal and providing additional services. Network equipment vendors generate revenues by selling or leasing equipments as well as provisioning equipment-related services. There are a range of revenue sources for content providers, they argue, such as subscription fee, pay-per-use, syndication agreement and airtime revenue sharing. Similar to variety of content provider revenue models, revenues of application providers may come from licensing, installation fees, hosting service, operation & maintenance service and consulting services. Payment agents, who are called charging and billing providers in our study, earn revenue streams from commission, payment platform development and operation. The major source of mobile operators revenue is subscription fee, together with potentials like whole selling network capacity to MVNOs (Mobile Virtual Network Operator) and operating portals. Internet service providers (ISP) charge subscription fees from customers and earn revenue from other ISPs based on their agreements. Regulators, which infrequently appear in revenue model analysis, levy license fees and various kinds of taxes. Despite that they do not conduct a further classification; the revenue models presented still offer a comprehensive coverage of legacy models. Thus it can be useful as an input of our

40 CHAPTER 3. INDUSTRIAL BACKGROUND 32 analysis Revenue models of Internet services There are numerous taxonomies of Internet business models. For the assessment of this thesis, the classification proposed by Afuah and Tucci [1]is chosen as a reference. One argument for this choice is its logical clarity and exhaustiveness, that is to say, it has distinctly and comprehensively illustrated most of the prevailing Internet business models. Another reason is their analysis is on the basis of revenue models, which is quite similar to our approach. Before digging into revenue models, Afuah and Tucci [1] identify eleven so called profit sites, of which the concept resembles actors in our analysis. They are respectively E-commerce, content aggregators, brokers/agents, market makers, service providers, backbone operators, ISPs, last mile, content creators, software suppliers and hardware suppliers. Seven sources of revenue for those profit sites are then depicted: Commission [1]: Third party service providers charge fee on the transactions conducted through them. Different variations of commission based model are further enumerated. 1) Buy/sell fulfillment, which is referred as transaction broker or online broker by other scholars, allows customers to conduct transactions (e.g. Scottrade, Orbitz). 2) Market exchange, which build up online marketplace playing a role like business facilitator (e.g. New View). 3) Business trading community, which set up a vertical site for people with specific interests to exchange information (e.g. Vertical Net). 4) Buyer aggregator, which gathers buyers together to gain bigger bargaining power over the sellers (e.g. Market Mile).

41 CHAPTER 3. INDUSTRIAL BACKGROUND 33 5) Distribution broker, which is essentially business-to business distributor (e.g. Grainger). 6) Virtual mall, which is an aggregate of links to merchants (e.g. Yahoo! Shopping). 7) Metamediary, which also provider clearing services in addition to a virtual mall (e.g. Amazon zshops). 8) Auction broker, which provides an auction place and charge fees to sellers (e.g. ebay). 9) Reverse auction, which provides an auction place where the sellers bid, that is a buyer name a product or service he intends to purchase and sellers bid for the deal (e.g. Priceline). 10) Classifieds, which is a vertical portal for trades of certain kinds of products or services (e.g. Apartments). 11) Search agent, which facilitates online shopping by scanning other sites and return customized results for customers enquiries (e.g. MySimon shopbots). 12) Bounty broker, which assists customers to search for hard-to-find products or services (e.g. BountyQuest). 13) Matchmaker, which helps business or individual to get what they need (e.g. iship). 14) Transaction broker, which is a third-party platform for transactions (e.g. PayPal). 15) Peer-to-peer content provider. Advertising [1]: The cost of service provision or content development is subsidized by advertising income. Users get services for free or for lower-than-cost prices, but in return they have to accept the advertisements via different methods. In order to achieve success in terms of online advertising, one site needs to attain either quantity of audiences or quality. Similar to traditional media like TV and newspaper, a bigger audience base one site has, the more attractive it is likely to be for advertisers. By quality of audiences, they mean the audiences exhibit common characteristics so that

42 CHAPTER 3. INDUSTRIAL BACKGROUND 34 they can be clearly aimed at by targeted ads. Advertising is further specified by the following variants: 1) Generalized portal, like Yahoo and MSN which cover a great range of content to attract visitors and thus build up big audience bases. 2) Personalized portal, customized by users themselves which consists of different modules according to their personal preferences. 3) Specialized portal, of which the content is all about a specific field (like Carmagazine). 4) Attention/incentive marketing, in which visitors can get monetary benefits from the website by viewing pages or clicking links. 5) Free model, in which visitors can get free goods or services by viewing ads. 6) Bargain discounter, in which the products are largely discounted in order to attract eyeballs. 7) Recommender system, where users share opinions about products or services. 8) Informediary registration model, in which users sign up services with certain personal information in exchange of free services (e.g. NYTimes). 9) Community provider, which provides a platform for online communities. Markup [1]: companies purchase goods from producers and resell them. They can pure online retailing businesses like Amazon which don t have physical retailing stores. Another variant, which is widely referred as click-and-mortar model, is that traditional merchants extend their businesses to the Internet. Production [1]: manufacturers utilize Internet as a marketing channel, for the purpose of directly reaching the customers without any intermediary or distributors in between. Five variants of this model are mentioned: 1) Manufacturer-direct, of which the most prominent example is probably Dell s

43 CHAPTER 3. INDUSTRIAL BACKGROUND 35 direct sell. 2) Content producer, in which firms directly sell their information goods. 3) E-procurement, namely companies move their procurement of goods or services online to reduce transaction cost. 4) Networked utility provider, which take advantage of network externality or two-sided market (e.g. Adobe). 5) Brand integrated content, which integrates ads into online content to moderate intrusiveness for viewers. Referral [1]: Sites earn commission by directing visitors to other websites. It could be pay-per-click or pay-per-sale, depending on the agreement between referral sites and destination sites. Subscription [1]: a subscription sells periodic access to a service, which can be either access to the Internet or value added services such as content. Fee-for-service [1]: users pay for the amount of services or products they consume, resembling the pay-per-use model in chapter 3.2. In addition to the seven revenues above, pricing models on the Internet are classified into six categories. They are fixed pricing, one-to-one bargaining, auction, reverse auction, barter and free Comparison of mobile and Internet services After introducing the background of both telecom and Internet industries, a comparison of the two offers constructive implications for the emergence of mobile Internet. In a

44 CHAPTER 3. INDUSTRIAL BACKGROUND 36 study of electronic commerce, Wu and Hisa [71] investigate common characteristics and differences between web-based commerce and mobile commerce. Their evaluation is broken down into two parts. The first step is an assessment from perspectives of three core elements, namely technology, content and service [71]. From the technological dimension, they stress that the lack of universal standard is a drawback of mobile services compared to the IP-based Internet. Cellular transmission is connection-based, message-oriented, device-dependent and has restricted bandwidth, imperfect coverage whereas Internet is the other way around. The strength of mobile services includes mobility, continuous presence and location awareness. As for the content, Internet services and mobile services are compared by development and distribution of content. Internet services are more information intensive and mostly pull-oriented while mobile content is typically push-oriented and limited to regional distribution. Moving on to the service dimension, mobile services enjoy an advantage in terms of location-awareness and personalization. Internet services, on the other hand, leave mobile services behind in functionality and usability perspective. They exhibit a wider range of services, an easy-to-use search, rich interaction and relatively better potential for integration. On the basis of the preceding findings, Wu and Hisa [71] take a further step to compare the business models of these two industries. The comparison consists of five aspects, viz. value proposition, targeted market, cost structure, profitability and value network. They firstly argue that high-speed, low-cost Internet enables rich communication that largely expels information asymmetry. In contrast, core value proposition is personalized, location-aware services can be delivered to end-users without temporal and spatial restrict. Compared to Internet services, mobile services are usually more precisely targeted at a certain market segment. As to the cost, infrastructure of

45 CHAPTER 3. INDUSTRIAL BACKGROUND 37 telecommunication is typically more expensive which indicates mobile information transmission is more costly. In the meanwhile, they claim that development cost of mobile service is always lower because of the simple presentation. About the profitability, the lean, direct-to-customer distribution channel of Internet services significantly cut down the cost while mobile services help Internet market expand by providing mobility. Lastly, the Internet value network exhibit a modular structure [58] where the telecommunication world has a vertical integrated industrial structure [56].

46 Chapter 4 Analysis of mobile Internet services After constructing the theoretical framework and exploring the business landscape of mobile Internet industry, the thesis will move on to look into the business logic behind the complicated mobile Internet world via applying the above-mentioned theories to it. This chapter starts with a description of the research methods used in this study and how they are executed. Then the second part will go on with an introduction of the objects to be probed in the case study. The next step is taken to analyze the value network of every selected service. At the end of this chapter, findings of advertisement-oriented services are generalized and an abstract model is developed. 4.1 Analysis framework Theories presented in chapter two are of closely relevance to this research. All of them provide perceptive implication and make reasonable contribution. However, none of them is perfect enough to describe the complexity of the business world of mobile Internet. An inadequacy of rationale makes it sensible to take inductive case study as the research method [14]. Case study as a methodology stands out at explaining a complicated contemporary phenomenon within its real-life context, when the boundary line between phenomenon and context is not of obviously evidence [72]. The intricacy and fast-changing inherence of mobile Internet industry also makes case study the best choice opposed to other research methods. Furthermore, we mainly focus on success stories in this research. That is to say, only those services regarded as successful ones are picked up for case study. In the rest of this subsection, the procedures of our case 38

47 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 39 study will be illustrated Case selection As defined in the first chapter, mobile Internet services in our research include data services used on handheld computing devices. Hence those non-data mobile services (e.g. SMS, call) and standalone applications (e.g. calendar, alarm clock) are excluded in spite of the fact they enjoy a widespread popularity. Detailed data on mobile application and service usage is gathered via a handset-based end-user research method. The dataset, which was collected in fall 2007, covers more than 500 panelists who are mostly Nokia smartphone users. Preliminary data processing was then conducted to adapt the raw data for an analysis of popularity of mobile Internet services. Usage of each service was measured from several dimensions, such as number of users and use frequency. We choose mobile Internet services for case studies based on two criteria. One is market penetration rate which is chiefly derived from number of tryers. A tryer is defined as a one-time buyer to distinguish from a customer [63]. A tryer turns into a customer only if his expectation is repeatedly fulfilled. The other criterion is customer retention rate which is educed from three factors, a) the number of users who tried the service multiple times, b) the number of users who repetitively used the service and c) the number of users who frequently used the service. We screened all the services to set up a threshold for each criterion. Services which at least surpass one threshold are then picked up for further analysis. Grounded in the preceding criteria, a set of services (table 4.1) are picked out for our

48 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 40 case study. The A, S, T in the last column indicate how the revenue model of each value network is interpreted by end-users, in other words the pricing model. Opera Mini is considered as using advertising model since its business logic has more similarities with advertising than with the other two. Fring has no evident business model at this moment. Being officially claimed it will be free, it will probably adopt advertising model or quasi-advertising model. The possibilities include, according to Fring [65], advertising and revenue sharing with its SIP provider partners. In the following analysis, Nokia Landmarks is regarded as part of Nokia Maps so that it won t be analyzed separately. Besides, the usage of these services is not exclusive. That is to say, user of one service may adopt one or more of the other services. Table 4.1 Selected services

49 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Within-case study Our case studies follow the guideline of configurational comparative research [46]. The idea is to implement both within-case study and cross-case study. The precondition of using this research method, notes Ragin, is existing theories are not well-formulated enough to present explicit hypothesis. It perfectly matches the situation here. During the study of selected cases, we keep it in mind that our goal is to make sense of both individual cases and sets of similar cases. Each case is firstly investigated separately to generate outcome for cross-case study.

50 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 42 Detailed information about the cases is primarily from publicly accessible sources, e.g. firms websites. In addition to official websites of these services, we also collect information from other sources, which are relatively more objective, so as to gain a comprehensive picture. It is comprised of magazine articles, analysts reports, blogs related to mobile Internet industry and so forth Cross-case study Combined with with-case study is a cross-case analysis to identify patterns. The idea is to look beyond the limited initial impression and look into evidence through multiple lenses [14]. Among all the cross-case study methods, a case comparison approach is selected for this research [14] [72]. The preparatory work for a comparison cross-case study is an adequate investigation of each individual case [72]. Then all the cases are split into different categories for the analysis of with-in group similarities together with cross-group differences. Along the analysis process, irrelevant variations need to be ignored to make the picture clear. Leveraging the results of within-case studies, we carry on with conducting a cross-case analysis of the business logic behind those relatively successful mobile Internet services. The cross-case study explores common characteristics of successful business models and tries to derive a general classification Value network model This framework is developed on the basis of Heikkinen s [76] definition of value exchange. In his work, all the incentives circulating in value networks, which are equivalent to the exchanged value here, are categorized into

51 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Goods (tangible, intangible, services) 2. Information (details of an actor, brand, reputation, trust, etc.) 3. Currency (money, tokens). The basic idea of Heikkinen s [76] framework is adopted while the definition is further adapted for the study of this thesis. Firstly, all three groups are renamed. Goods are changed into services since tangible goods can hardly be seen while currency is relabeled as monetary benefit. Information is replaced by intangible benefit because one key element that we intend to include in this group is attention and it is better to be regarded as a kind of intangible benefit instead of information. Secondly, two sub-categories are derived from the first group. One is core service which characterizes the central benefit delivered to end-users. It is substantially the principal purpose for which a user chooses the service. The other is enabler service which facilitates and enhances the core service. One important variant of enabler service in this analysis is distribution channel. Improved efficiency of channels has a positive impact on the performance of core services. Thirdly, in addition to attention and loyalty, another element (i.e. information) is included in intangible benefit. In this scenario, information stands for details of an actor, more specifically it denotes background and preferences of end-users. For example, service providers get hold of personal information or usage information of their users. It could make their ad places more valuable if they are able to take advantage of this information to send more precisely targeted ads. All in all, we distinguish among three types of value exchange. Each of them is

52 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 44 assigned a symbol which will represent the corresponding value exchange in the subsequent value network analysis: Services & Goods S1: core service (central benefit for users) S2: enabler service (facilitating core service) Monetary benefit: M Intangible benefit I1: Attention I2: Loyalty (incl. brand recognition, reputation, etc.) I3: Information (actors background, preference, etc.) Let s take Google search engine as an example to shortly explain how those symbols will be used. Figure 4.1 Value exchange of Google search In figure 4.1, the diagram on the left indicates that Google provides free search service for users in exchange of their attention, loyalty and information. The one on the right, on the other hand, shows that Google sells the attention and information of users (i.e. ad place) to ad agencies in exchange of money. The former value exchange adopts advertising model while the latter one embraces transaction model.

53 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Description of selected services and actors Selected services Visual Radio Visual radio is an application from Nokia to enrich audiences interaction with radio programs. It is intended to improve experience of traditional radio broadcast by bringing synchronized images and texts to users. With radio built-in handsets, users still listen to programs through regular analog radio. In addition to that, the application provides a visual channel with more information and opportunities of interaction via cellular network. While listening to radio, not only can users access more detail information about the programs such as the artist or lyrics of the music playing on air, but they can also participate into other interactive activities like audience polls and digital content download. Numerous services can be materialized by visual radio. For instance, users may get gossip information about celebrities, purchase concert tickets or CDs while hearing a delightful song, or have a look at the missing part if listening to news from halfway through, etc. Nokia Maps Nokia Maps is a service for selected Nokia handsets. It offers free mapping and routing services for mobile device users to find the way to their destinations. Besides, it provides information about points of interest like tourist interest, restaurant and hotels. In addition to these free services, premium services including city guide and voice navigation are available for handsets with GPS capabilities.

54 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 46 Lifeblog Lifeblog is a multimedia diary application provided by Nokia. It automatically organizes all the pictures, videos, text messages and MMSs on one handset into a single timeline. Moreover, with Lifeblog PC client, those contents can be easily transferred and synchronized between PC and mobile phones. Another important feature of Lifeblog is handset users can upload photos and videos to their blogs without any difficulty. This makes the capturing life memories and share of experiences a lot easier. Nokia Search Nokia search consists of four applications, namely Local Search, Web & Image Search, My Content Search and Media Roaming, among which the last two are standalone applications so that they will be excluded in our study. Nokia Local Search is a context-specific search service offered by Nokia. It enables users to find the restaurants, pharmacies or other kinds of services that match the geographic criteria defined by them. For instance, when one has a headache and needs some aspirin, he probably is anxious to know where the nearest pharmacy is. Nokia Local Search returns the user with the phone number of the closest pharmacy together with its location on Nokia Maps. The user can choose either to call them or use Nokia Maps to navigate to it. Web & Image Search provides mobile access to world leading Internet search engines including Google, Yahoo, Live Search and Baidu.

55 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 47 Navicore Navicore is a GPS navigation service from a Finnish corporation called Navicore. It is a software solution which leverages on the hardware capability of mobile devices. The service package includes pre-installed maps, route calculation as well as voice and visual navigation. User may try it free for 7 days, and beyond that purchase of license is necessary. Opera Mini Mini, which is from Opera, is a free web browser mainly aims for handsets. The lightweight browser has a size of only 100kb which can perfectly fit into a big range of low-end and middle-end handsets. Thanks to its proxy server and client-server architecture, Opera Mini is able to bring equally pleasant-someone argue it is even better-user experience as the mainstream smartphone browsers. Opera Mobile Opera Mobile is another web browser from Opera. Unlike Mini, it reformats web pages to fit into handset screens instead of having servers to pre-process these pages. Opera Mobile has a free trial period of one month, but beyond that users need to pay. Nokia Channels Labeled as a media service, Nokia Channels renders users access to newspapers and other media around the world. It is a syndication tool to get updated content, either via WLAN or GSM and then users can read them either online or offline.

56 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 48 Google Mail Google Mail is a free client for handsets users to access their Gmail accounts. It enables users to go to their accounts without opening a web browser. With Google Mail, users could enjoy enhanced user experience of browsing and writing s. Google Maps Google Maps (precisely called Google Maps for mobile) is a mobile client of Google s mapping service. It has most features of its PC application, for instance satellite image, driving navigation, local businesses and real time traffic information. Nokia WidSets Widest is a runtime mobile service for users to keep track of updates to websites they often visit. It uses widgets, which are mini-applications, to maximize the use of screen space. Whenever there is any update with the content, which could be news, blogs, s, music, pictures, etc, Widsets pushes it to the user s handset. With Widsets, users don t have to visit web sites or check client applications regularly to fetch the content. Fring Firing allows handset users to access their instant messengers on the move. With Fring, users can make free calls and live chat with their friends on instant messengers via the data link. Not only does it integrate all the contacts into one buddy list, but it also offers PC benefits, such as contact availability indicator, to handsets [75].

57 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 49 Nokia Landmarks Landmarks is a sub-function of Nokia Maps. It allows users to store points of interests for later use such as navigation to these places. Furthermore, users can also share the location of specific places with their friends. Elisa Mobi Elisa Mobi is a mobile service package from Finnish mobile operator Elisa. It is pre-installed in the mobile phones provided by Elisa. With the Elisa Mobi menu, users are empowered to access services at one touch. It contains a big variety of services such as mobile TV, music, video call and different Internet services. TomTom TomTom stands for TomTom Mobile which is a software solution designed for smartphone. It is a handset compatible version of the company s leading GPS navigation product. In addition to navigation, it has several premium services such as real-time traffic and weather information. Users need to purchase the software and maps in order to use the service. Nokia Navigation As another sub-function of Maps, Navigation has a walk mode and a drive mode. Walk mode is designed for pedestrians, offering them turn-by-turn visual navigation and information of surroundings. Drive mode is a navigation system for car drivers. Both the modes require purchase of license for legitimate use.

58 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 50 Nokia Internet Radio Internet Radio is a radio broadcast service transmitted through Internet. It provides free access to hundreds of radio stations around the world. MSN MSN is a free instant messaging client from Microsoft. It allows user to contact with their friends in real time with a wide range of rich content like texts, icon, voice and video. Nokia Photo Print Nokia Photo Print is a fully integrated mobile printing service in Europe, allowing users to order photo prints directly from their compatible Nokia devices and have the prints delivered either to their home or their friends. To use Photo Print service, users simply select the picture in the gallery application of Nokia handsets, and the enter address and other required information. Then the order is handled by a Nokia partner and prints are delivered to the recipients. Nokia Music store Announced in late 2007, Nokia Music Store sells media files through both cellular network and Internet. That is to say, not only does its web-based, itune-store-like service enable users to download songs to their PCs, but it also offers over-the-air digital music purchase and downloads with handsets. In addition, two-way synchronization facilitates sharing music between mobile devices and host PCs.

59 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Industrial actors Before unfolding the analysis of value networks, it is important to have a fairly unambiguous definition of several industrial actors that appear in this research. Mobile operator Mobile operators in this thesis are the firms that provide mobile network subscription for customers. They incorporate mobile network operators which also own the infrastructure and MVNOs (Mobile Network Virtual Operator). MVNOs purchase network capacity from network operators and sell it to subscribers. Mobile operator is involved in value networks of all the mobile Internet services since connectivity is vitally important for them. Advertiser Advertisers place advertisement in services to carry out advertising campaigns, using services as an advertising media. They show up in all advertisement oriented services. In this study, advertisers include both advertisement agencies such as DoubleClick and companies which directly purchase ad places from mobile Internet service providers. Search partner Search partners are the partner companies which offer search engine service for users. In this study, they only appear in the value network of Nokia Search, specifically indicating Nokia s search partners including Google, Yahoo and Windows Live.

60 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 52 Content publisher Content publishers provide content for consumers, which is accessible from mobile devices. They gather content together from various sources and normalize structure. Content publishers are either the owner of the content or the one who aggregate content created by third parties. In this thesis, content publishers specifically refer to the sites provides tailored contents for Opera Mini users. Merchant Merchants referred in the location-based services are the businesses which physically deal with customers. They include retail stores, restaurants, banks, and so on. Merchants can be found in the value networks of those navigation services which provide information about points of interest. 4.3 Descriptive analysis This subsection presents general descriptive analysis of selected mobile Internet services. Before digging into the value network analysis, several basic properties of these services have already been identified, not least of which are the revenue model and the service provision company. In the following part of this subsection, descriptive results are reported from these two perspectives. Among all the services (table 4.1), Visual Radio has built up a commanding lead in terms of market penetration rate while Nokia Maps is slightly ahead of second-place Navicore regarding the number of users. Nokia shows a strong presence on its homeland. Over half of the selected services are orchestrated by the handset giant.

61 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Comparison of revenue models Table 4.2 Comparison by categories of revenue model We sort out all the selected services according to the revenue model each of them uses (table 4.2). Among other things, the advertising category outnumbers the other two, transaction and subscription, in most respects. Firstly, there are 13 services that embrace advertising model out of the top twenty selected mobile Internet services. Secondly, on the subject of market penetration rate (i.e. number of tryers), the advertising category overwhelmingly outshines. Top four spots on the ranking list are taken by advertisement-oriented services. Not only is the total number of tryer of advertising services as more than three times as the sum of the other two groups, but they also win on an average level, 48.4 versus 30.0 of subscription and 22.7 of transaction. Thirdly, due to a large tryer base, the advertising set also exceeds in terms of the number of users. Nevertheless, the predominance is not as significant as it is on

62 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 54 the topic of penetration rate. A potential explanation for this relatively less prevalence is services using advertising model mostly have a lower barrier of entry. People can try out these services without any cost or consequences. Yet not all the services could offer satisfactory user experience. According to Weinstein & Johnson [63], repeated fulfillment of customers expectation results in high customer loyalty, Loyalty in turn % brings about enhanced business performance and increased customer retention rate. Figure 4.2 Comparison of mean customer contention rates 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% Advertising Transaction Subscription Tried % Multi-trial % repetitively Used % Figure 4.2 shows a comparison of mean customer contention rates of three categories. Subscription tops in this regard. On average, 42.84% of its one-time tryers turn into multi-trial users, of whom more than seventy percent end up with using the service repetitively. In contrast, multi-trial users and repetitive users take up only 11.90% and 10.32% of tryers of transaction services respectively. Advertising is in between, which is moderately left behind by subscription but transcends transaction with a big margin. The outcome demonstrates the advantage of subscription model, namely reduction of risk. Not only do users additionally attached to the services, but they also pay the bill in advance which helps secure the financial standing of firms providing services.

63 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Comparison of company market shares Figure 4.3 Comparison of number of tryers and users by company a. Number of tryers by company b. Number of users by company Among other things, the diagrams (Figure 4.3a, Figure 4.3b) show Nokia s strong influence in its home country. Taking advantage of its strong power in Finnish telecom industry, Nokia secures a dominant position in the service domain. Internet service firms, including Opera, Google and Microsoft, leverage on their experience of serving end-users and manage to gain notable market shares. Navigation service, which is widely considered as a promising money cow, helps navigation service providers Navicore and TomTom earn a place in the rank. 4.4 Value network study After the service selection, each service was studied separately. During the case studies, we figured out that value networks of some services exhibit high similarity. In the rest of this subchapter, therefore, not all the analysis will be presented in equal detail. Among advertising model services, seven of them are exhaustively elaborated (Visual

64 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 56 Radio, Nokia Maps, Nokia search, Internet Radio, Google Maps, Opera Mini and MSN). Value networks of Nokia Channels and Widsets are going to be briefly described thanks to their likeness with that of Internet Radio. Likewise, value networks of Google mail and Lifeblog will be shortly explained because they closely resemble the value network of MSN. Up to date, it is still not crystal clear what the business model of Fring is. Not only did Fring people claim it would be free, but they don t seem to have billing structure so far. In an interview, Avi Shechter, co-founder and CEO of fring, revealed that the sources of income could be either revenue sharing with SIP partners and third-party application providers or advertising [65]. The former one bears a resemblance to Visual Radio while the latter one takes after Internet Radio. Both of the possibilities are taken into account but will not be illustrated in the value network study part. Landmark will be considered as part of Nokia Maps, and thus it won t be explained individually Visual Radio Figure 4.4 Value network of Visual Radio

65 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 57 The capital A, S, T in blue represent three revenue models respectively, namely advertising, subscription and transaction. Every place where is one of those letters indicates the revenue (or pricing) model between two parties which have direct relationship with each other. For instance, end users subscribe to mobile services while Nokia adopts advertising model for service provision to end-customers. Those solid lines stand for service flows. Nokia delivers the core service to end-users while mobile operators make the access to network available. The Nokia service could undoubtedly, no matter how much, increase the data traffic. It, in some sense, provides a distribution channel for mobile operators to sell its network capacity. On the other hand, the support to this Nokia service given by mobile operators (at least no boycott) offers subscribers with accessibility to Visual Radio which can be regarded as a distribution channel for Visual radio. The technical requirements are sent to Nokia by mobile operators for the service development. Besides, the rich contents and ads in Visual Radio are provided by radio stations. In this sense, the Nokia service supplies a distribution channel for the multimedia content and value-added services of radio stations as well.

66 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 58 The broken lines illustrate flows of intangible benefits. Customers give their attention as well as brand loyalty to Nokia as the exchange of free service. The attention flows through Nokia to radio stations till they exchange it with ad agencies for monetary benefit. As agreed with Nokia, radio stations give a slice of the revenue generated out of Visual Radio with the handset giant. In addition, radio station can acquire knowledge (i.e. so-called information in this analysis) about users of visual radio service. A user shows his preference or interest while listening to a certain program. Then targeted ads can be delivered through visual radio to him. The dotted lines represent flows of monetary benefits. They come up in a majority of relationships where transaction model or subscription model takes place. Customers provide payment for sellers in exchange of services or goods Internet Radio Figure 4.5 Value network of Internet Radio

67 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 59 The business logic behind Internet radio partly resembles that of Visual radio. The major difference is, instead of passing attention to radio stations for them to sell ads, Nokia deals with ad agency itself in this case. Most radio stations seem to be willing to allow Nokia to include their broadcast. They broaden their distribution channels without any additional cost in an era when popularity of radio has been slipping Nokia Maps Currently Maps doesn t generate any revenue for Nokia if the navigation service is not taken into account. According to Nokia, however, there are three potential sources of revenue for its location based services besides navigation. It could serve as an advertising channel a la Google Maps or charge referral fees from businesses which connect to Nokia users through the handset vendor. Another possibility is to sell premium contents (e.g. GPS waypoint coordinates, point of interest location information) to Maps users. Figure 4.6 Value network A of Nokia Maps

68 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 60 The first potential revenue source of Nokia Maps reminds us of the referral-based model [1] used by Internet firms, which was introduced in chapter 3.3. Nokia steers its Maps users to its partner merchants. In exchange, the handset giant takes a portion of the eventual revenue sales but it could also be a flat fee. Figure 4.7 Value network B of Nokia Maps

69 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 61 Drawing on the information about users physical location, Nokia makes location-aware advertisement possible which could more precisely target customers (figure 4.7). Figure 4.8 Value network C of Nokia Maps Users simply purchase premium contents that they desire to get from Nokia (figure 4.8) Nokia Search Figure 4.9 Value network of Nokia Search

70 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 62 Figure 4.9 illustrates the value network of Web & Image Search. It takes after value network of Visual Radio (Figure 4.4). Nokia provides users with mobile access to its search partners. It acts as an additional distribution channel for these search engines. Local Search, which is part of Nokia s location based services, shares the same business logic as Maps. (Figure 4.6, 4.7 and 4.8) Opera Mini There are at least four ways Opera makes money out of its Mini. The first one, and potentially the biggest cash cow for Opera, is the revenue-share agreement with mobile operators (Figure 4.10). Opera grants mobile operators to pre-install Mini into their bundled handsets. According to Opera, subscriptions with Mini installed handset are highly likely to generate more data traffic than the ones without. As a result, Opera gets a slice of additional revenue resulted from Mini. T-mobile Germany was the first one

71 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 63 pioneered to launch subscription with Mini bundled and reported a 119% surge in data ARPU. Figure 4.10 Value network A of Opera Mini Secondly, Opera earns revenue by selling proxy servers to content providers (Figure 4.11). Opera offers free browsers with no-compromised quality to build up a massive customer base. Then it starts selling proxy servers for content providers to tailor their content into a more Mini-friendly format. Figure 4.11 Value network B of Opera Mini

72 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 64 The third approach, which provides a fairly solid source of income for Opera, comes after its compensation deal with its partners. The Mini browser comes with several built-in bookmarks and a search box with Google as the default search engine (there is a drop-down where users can choose from a few other search engines). Google and other partners that take up places in the bookmark pay for increased popularity Opera Mini brings them. Mozilla foundation, which offers one of the most powerful Mini rivalries Firefox, has a similar strategy of having Google in the default search box. According to Mozilla, Google accounts for 85 percent of its revenue in 2006 [73]. The last money-making way is to sell customized Mini browser. For instance, Opera has established partnership with ebay and launched an ebay version of Mini in Germany. With the tailored Mini, German ebay users can easily bid and sell via their handsets. The one-click access to ebay content and function also frees them from hassle and saves their time.

73 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES MSN Figure 4.12 Value network of MSN Microsoft essentially adopts advertising model for its MSN messenger (Figure 4.12). It has been giving away the MSN messenger-or via bundling it with its operating system-to accumulate a huge base of users. Currently MSN messenger is one of the most used instant messaging software in the world. The ways of placing ads in MSN are rather versatile [74]. In the beginning, the advertisement in MSN messenger was limited to a banner at the bottom of the program window. Then they added text links, which may show up in the conversation window and sharing folders, as well as tabs by the side of program window. After that, MSN messenger released so-called branded experience advertising in which they customize backgrounds, icons, winks, etc. with advertising themes included. While introducing the feature of video conversation, Microsoft enabled advertisers to place video ads. The ads are displayed in the meantime while the connection is being established.

74 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Google Maps Figure 4.13 Value network of Google Maps The business logic of Google maps is more complicated than the preceding ones. It is like a hybrid of several business models (Figure 4.13). Google maps have open APIs for mash-up sites to integrate Maps into their services. Google s openness to the use of its maps does have limits, though. Once a mash-up turns into a large-scale commercial enterprise, Google looks to share in the revenue. Google seems to be pursuing revenue-sharing deals with sites that make money from Google Maps.

75 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 67 On the other hand, Google has Adsense in Google Maps to sell ads, which applies the generic two-sided business model Lifeblog Appendix A.1 shows the ways Lifeblog generates revenue for Nokia. On one hand, it is a blog service itself which empowers Nokia to sell ad places to advertisers. On the other hand, Lifeblog provides additional access to other blog providers, which opens up revenue sharing possibility Navicore, Nokia Navigation Navicore (Appendix A.2) and Navigation of Nokia Maps (Appendix A.3) share exactly the same revenue model except for slight differences with respect to pricing and distribution channel. Users need to purchase licenses so as to use the service for certain period of time Opera Mobile Opera Mobile has two types of business models (Appendix A.4). One is to directly serve consumers by proving them free trial-version of the browser. After a trial period, users need to purchase the browser which is a one-time payment without any subsequent cost. The other way is bundling with handsets from the major handset vendors and operators. Opera charges a license fee from each mobile phone shipped to the end-users.

76 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Nokia Channels Appendix A.5 depicts the business model of Nokia Channels which resembles that of Nokia Internet Radio. Channels serves like a portal of all the contents Google Mail Appendix A.6 illustrates the business logic behind Google Mail. Google sells ad places (AdSense) beside users s to advertisers Nokia WidSets Appendix A.7 shows the ways Nokia makes money out of WidSets. For service providers, they may choose between directing the traffic to their mobile Internet sites and working with Nokia to make their own widgets with supporting ads. As an exchange, they share the revenue generated via WidSets with Nokia. An alternative for Nokia is they directly sell the ads on WidSets to advertisers Elisa Mobi Consumers subscribe to the services listed in Elisa Mobi. If it is a white-labeled service or a third-party service, Elisa is likely to provide its partners with monetary compensation or share revenue with them. (Appendix A.8) TomTom Unlike Navicore and Navigation of Nokia Maps, TomTom (Appendix A.9) only charges users fees for the initial purchase of the software and map data. Unless users

77 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 69 choose one of the premium services like real-time traffic and weather information, no additional bill will be sent to them Nokia Photo Print End users pay for Nokia Photo Print service on the basis of the number of prints. Logistics are outsourced to different local delivery companies depending on the region. (Appendix A.10) Nokia Music store Consumers have the choice to purchase either by tracks or by album. The way Nokia addresses copy right issue with labels varies from case to case and is not unveiled in certain cases. It could be revenue sharing according to the amount of music having been sold or a license fee for each Nokia phone shipped. (Appendix A.11) 4.5 Classification of advertising models Almost two thirds of the selected services are found to adopt the advertising model. It implies that the advertising model may turn into the mainstream in the future. In this study, our intention is to figure out the business logic behind mass market. Besides, the number of case studies should be big enough to guarantee a thorough and systematical cross-case investigation [72]. For both of these reasons, cross-case study among advertising model services became our natural choice. The concept of two-sided market can apply to most advertisement-oriented services in mobile Internet market. Thus abstract models are conceptualized based on the theory of two-sided market and compared against each other. Finally four types of advertising models are generalized. Each of the four variants emerged from our research i.e. two-sided market, double two-sided market, three-sided market and hybrid, is discussed below.

78 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES Two-sided market The first model, two-sided market, represents the generic two-sided market which is described in chapter 2.3. It depicts scenarios where only three sets of actors are involved in the core value exchange. Among the three, one of them plays a role of platform provider which separately maintains a relationship with each of the other two. Platforms are defined as services that attract groups of users in two-sided networks [15]. Grounding on basic ideas of the original theory, three variants emerged from our research. Figure 4.14 Two-sided market variant A The first one specifically reflects value exchange characteristic of advertisement-oriented services. Platform providers subsidize services so as to augment the user base, no matter it is targeted at the mass audience or aims for a niche like those vertical portals do. In the meantime, they are able to acquire attention from end-users, and then sell it to advertisers for cash. Both Google mail and MSN use this strategy. Figure 4.15 Two-sided market variant B

79 CHAPTER 4. ANALYSIS OF MOBILE INTERNET SERVICES 71 In the second variant model, platform providers also offer subsidization to side A, which are typically customers of both platform providers and side B. Unlike the previous model, platform providers pursue loyalty (i.e. stickiness) from side A instead of attention. On the other hand, they sell to side B at premium prices because they help side B make their products or services more appealing to side A. It is a classical example how Acrobat prices PDF reader and writer. Video game consoles like Xbox and Playstation typify this model as well. Amongst the selected services, the first model of Opera Mini can be characterized into this category. Figure 4.16 Two-sided market variant C The third variant reveals cases where platform providers subsidize side B which directly or indirectly serves side A as customers via the platforms. Products of side B add value to the platform. The additional money side A pays makes up for the subsidization to side B. Windows operating system from Microsoft embodies this model. Development tools are offered at very low prices in order to motivate developers to work out more applications for the platform. In out analysis, the second model of Opera Mini belongs to this class amid all the selected services.