1 Big data and the democratisation of decisions A report from the Economist Intelligence Unit Sponsored by
2 Contents Preface Introduction 4 Orchestrating the big data strategy 4 Making cross-functional teams big data strategy architects 6 Leveraging key data and departments 8 A paradigm shift 9 Appendix: survey results 0
3 Preface Big data and the democratisation of decisions is an Economist Intelligence Unit research report, sponsored by Alteryx. Our thanks go to all survey respondents and interviewees for their time and insight. The author was Carolyn Whelan and the editor was Riva Richmond. The findings and views expressed in the report do not necessarily reflect the views of the sponsor. October 0 About the survey In August 0 the Economist Intelligence Unit conducted a survey sponsored by Alteryx of 4 global executives to gauge their perceptions of big data adoption. Fifty-three percent of respondents are board members or C-suite executives, including 66 CEOs, presidents or managing directors. Those polled are based in North America (4%), the Asia-Pacific region (7%), Western Europe (%), the Middle East and Africa (6%), Latin America (%) and Eastern Europe (4%). Half of executives work for companies with revenue that exceeds US$00m. Executives hail from 8 sectors and represent 4 functional roles, including general management (0%), strategy and business development (8%), finance (7%) and marketing and sales (0%).
4 Introduction Nearly half (48%) of respondents say someone other than the chief information officer drives big data processes. 70% say a positive outcome resulted from their last major decision in which data played a pivotal role, while 4% say more big data would have helped. Much as the Internet upended industry during the last two decades, big data s potential to provide companies with deeper insight into their businesses seems destined to generate both upheaval and endless opportunity. The tumble and roll should only increase as the data heap grows. Abundant data and the technology to crunch them are now enabling more firms, large and small, to extract value from fresh insights. The data swell is fed by billions of transactions, web clicks and sensor readings, and the buzzing of smartphones and social media sites. Thanks to cheap servers for storing data and affordable supercomputers, algorithms and visualisation tools for mining it, companies around the world are better able to spot and act on new trends and intelligence. This expansion of corporate access to big data is now being met by a democratisation of which employees inside companies are able to tap data, draw lessons and make business decisions. The tyranny of technology and data specialists is breaking down. Now, players from many departments are harnessing data to make better tactical decisions about how to respond to emerging trends, find new business opportunities, retain more customers and goose their companies bottom lines. According to a survey of 4 global executives, conducted by the Economist Intelligence Unit and sponsored by Alteryx, nearly half (48%) of respondents say someone other than the chief information officer (CIO) drives big data processes. This trend should accelerate as corporations work to capture more value from their information assets. Over three-quarters of respondents say businesses must empower more employees to access and make decisions using big data. [We] rely on data for most things, says one respondent from the publishing industry. Our biggest challenge is to get info in the hands of those who can use it to grow the business. Survey respondents believe that tapping big data will aid decision-making around seizing market opportunities (66%), holding on to customers (%), competing with rivals more effectively (4%) and boosting financial performance (%). They also believe that allowing more workers to harness big data will help them make better and faster decisions (6%), illuminate business opportunities that were not previously apparent (4%) and identify and exploit opportunities more quickly (7%). Tellingly, the data most available to executives internal information are also what they most value, our research shows. To extract the most value from big data, companies first order of business may be to eliminate data silos between departments and find better ways to work together.
5 Orchestrating the big data strategy In the Don Draper era, the person who could convince someone of something was in charge. In a data-driven world, whoever can ask the right questions is the leader. Alistair Croll, analyst at Bitcurrent Some large organisations have long leveraged giant data sets. For decades household names like Wal-Mart, Tesco and Kroger and successful firms in industries like fast food, financial services and healthcare have sifted through transactional information, US Census Bureau records, company s and even genomic patterns to help inform decisions. This has helped guide actions around stocking shelves and opening new stores, battling fraud and gauging the probability of success or failure for a new drug. Many of these trailblazers offer useful lessons. The most important is to start with a highly focused problem, question or business priority. From there, managers empower teams to get answers and apply findings. For example, a question like Where do people walk within a store? can be answered with the help of facial-recognition software. Then a marketing team can improve product displays and selections. At most companies, democratisation of datadriven decision-making is largely aspirational. Though our survey respondents acknowledge the importance of big data and its link to driving better performance, only 7% of respondents consider themselves leaders in this area. Forty percent say their organisations are adequate at the collection and analysis of big data, while a startling 4% say they are somewhat or completely inadequate. It s about access but also about understanding, says one survey respondent. We all need to get better at amalgamating and analysing disparate data sources. Often it s the hammer looking for the nail, says Venkat Rao, a research associate with the Leading Edge Forum at Computer Sciences Corp, a technology systems integrator. Executives try to bang out insights from a raft of data rather than let business problems or priorities dictate the research. Most people are trying to wrap their heads around this, he says. Many firms are hamstrung by the lack of a cohesive big data strategy. Costly legacy technology infrastructures, rigid technology teams and old ways of working that prize intuition and experience over data-driven decisions are also common obstacles. In the Don Draper era, the person who could convince someone of something was in charge. In a data-driven world, whoever can ask the right questions is the leader, explains Alistair Croll, an analyst at Solve for Interesting, a firm that studies emerging technologies. Data make everything very much a meritocracy. It s OK to be wrong if it s quick and you learn each time. Organisations that can t increase the speed at which they learn will die. The cycle time is scary. Despite the imperative to act, investment is lagging. Nearly half (47%) of respondents say they do not expect to invest in expanding their ability to use big data in decision-making in the next three years, with financial constraints cited as the biggest barrier (7%). Larger firms are a step ahead, perhaps because of deeper pockets. Survey respondents who work for companies with revenue of US$bn-0bn are more likely (9%) to consider 4
6 Taking big data across Dell: leadership matters Four years ago, computer giant Dell asked executives worldwide to make their data available to other business units. But there was resistance. Executives were reluctant to share proprietary information that had been costly to collect and process, and little sharing occurred. That changed quickly when Chief Executive Michael Dell took charge, illustrating the importance of executive leadership in creating an environment that gives employees autonomy to leverage big data for better decision-making. He required executives to supply data and issue regular reports. Next, he embedded data scientists in each business unit to decentralise access to the data and action on new insights. Disparate systems didn t speak the same language, recalls Rob D Schmidt, executive director for business intelligence in the CIO s office. We had to break down barriers around data ownership [and] technology and define a common model. Only when Michael [Dell] required everyone to standardise on a single reporting format did we get there. Now we are very much aligned. Today, Dell pushes data from businesses and regions around the globe onto a common platform. It collects and acts on intelligence about the activities of prospective buyers on its global websites to assess tastes and response times. It watches for intruders, to detect and stop security breaches. Some 9,000 servers send regular reports to technicians who fix problems. And it tracks key social media influencers who are applauding or complaining about products or services, so Dell can respond in real time. Revenue and efficiency gains have totalled many millions of dollars, Mr Schmidt says. And efforts to extract more value from additional data types are under way. themselves industry leaders; 86% of this group believes it is important to empower employees with more data to make better business decisions. Corporate culture can also get in the way. Interestingly, North American respondents consider culture to be the top obstacle to sharing big data more widely (it ranked second globally). Some executives fret that employees will be overwhelmed by a tsunami of messy information that may confuse or distract already overloaded departments. After all, the volume, variety and velocity of information are both empowering and intimidating. A lack of consensus at leadership levels can impede the creation of a cohesive overall strategy, as can concerns about privacy and security breaches. Even firms willing to channel funds into better big data capabilities struggle with talent acquisition. Data specialists are in short supply and often lack industry expertise. There are not enough foot soldiers to fight the war, says CSC s Mr Rao. Often, rank-and-file employees are stepping in, on their own, to do battle. As with the Arab Spring, much of the democratisation of data use is emerging from the grassroots, as employees seeking ways to do their jobs better, find and mine data with tools of their own. Marketing departments are going rogue, says Solve for Interesting s Mr Croll, describing shadow data-collection systems that use tools like Google Analytics. To avoid duplication, use data efficiently and conserve resources, the C-suite would be wise to commit to a thoughtful big data strategy and to leverage this entrepreneurial mindset across the enterprise.
7 Making cross-functional teams big data strategy architects Pioneers in the use of big data task mid-level manager teams with strategy development that is rooted in unanswered questions and business problems. Marcia Tal, founder of consultancy Tal Solutions and a former executive at Citi who was responsible for building and leading the company s global decision-management system, recommends creating cross-functional teams that use data and analytics to develop products and services that solve problems. Together, they can prioritise their most pressing data needs while focusing on data that are accurate, relevant and actionable. Based on the collective judgment of the team, data, analytic findings and insights can be applied to business decisions, she says. But to get there, vision, strategy and investment are required. You need connectors people that can translate the business strategies, intentions and opportunities to the particular program initiatives, Ms Tal explains. Corporations need to blend art and science, employing creative data artists that draw the picture, which data scientists have discovered. In a structured but challenged department that needs to meet its numbers, it s EMI s success by numbers EMI Music wants even its most junior employees to be able to make quick decisions about how to best promote its artists around the world. It s their job to combine their skills and judgment with the vision of the artist and the cold, hard data to come up with a plan, says David Boyle, senior vicepresident of insight at EMI Music. At a company like EMI, relying on analysts or data experts to make decisions using data doesn t scale. Employees first task is to identify the audience for an artist or piece of music using EMI s own consumer research of a million in-depth interviews with consumers around the world. Then, ten web-based applications accessible to all on PCs, and ipads during meetings help them make decisions about products, pricing, sales channels and marketing approaches. For instance, if an employee promoting The Beatles in France decides the best targets are cash-strapped young people and gadget lovers, she might offer digital products and promote them on social networks. Mr Boyle says his top challenge has been to provide employees with high-level information at a glance and the ability to go deep to answer tough questions. I want quick, but I want complex. And that s incredibly difficult, he says. One trick: colour coding. For instance, blue highlights excellence and a lack of blue signals the need to dig deeper. Finding a balance has paid off. Many employees make well-supported data-based decisions daily and with confidence and speed. Partners get on board more quickly, product and pricing choices are smarter, and marketing is more focused and effective and EMI artists are better served. 6
8 hard for people to get their arms around that. These are difficult discussions. Are you learning as fast as the world is changing? she asks. Because traditional data and management structures are not usually set up to create value across business entities. For these transformational shifts to take root, we need leaders to recognise the importance of analytic business leaders within an organisation, she says. When formulating a big data strategy, tapping business-unit managers who are already capturing value from actionable insights may also yield useful guidance about key issues like risks, pricing and customer behaviour. Democracy, of course, is on a continuum, and leadership styles and cultures shape how data are shared. These styles can range from democratic to autocratic, says Jeffrey Stanton, a big data expert at Syracuse University. Most organisations work more effectively when senior management articulates a vision for how data should be used, he says. But that vision should be compatible with the organisation s mission and culture. 7
9 Leveraging key data and departments Internal data sit high atop our survey respondents ranking of the most beneficial big data types (6%). Significantly, fewer respondents cite external website content as beneficial (0%), and even fewer cite much-hyped social media content (9%), though it remains early days for social media. Much of today s data-driven decision-making also incorporates targeted, discrete data revealing market dynamics bought from external data aggregators. The sweet spot for sharing and spreading useful information seems to be in the product-line area. There, executives on the front lines are combining a keen understanding of their customers with internal big data and relevant external information about broader product trends purchased for relatively small sums. The combination means more effective decisions about which products to offer. Indeed, our research shows that analysing big data sets, whether from internal databases or culled from the Internet, are most valuable when viewed in the context that outside market and customer data can provide. More than half of our survey respondents believe there is particular value in analysing internal data alongside competitive intelligence, customer-segment information and market-condition models that companies normally purchase. According to our survey, the job functions that stand to benefit the most from big data analysis capabilities are customer service, marketing, strategy and business development, general management, and information and research. Customer-service and marketing departments have long profiled current and potential customers by dissecting demographic, relational and contextual data. So it comes as no surprise that over half of our respondents single out customer retention and spotting new market opportunities as business decisions that most benefit from tapping big data. The empowerment of employees to access and analyse data must be accompanied by freedom to act on their insights. Data empowerment is completely toothless unless you combine it with empowerment to take risks, Mr Rao says. And of course, companies need to protect sensitive data. Not everyone should be nor has the skills to be stewards, observers or analysts of big data, cautions Ms Tal. This requires very specialised skills. Companies should take steps to ensure that they control who sees and uses what information. 8
10 4 A paradigm shift Unleashing the power of big data enables innovative executives and employees to find new market niches, tweak product offerings, reduce inventory and boost profit margins. But capturing value from actionable insights requires a paradigm shift in culture and execution. Through a cohesive, integrated strategy that is backed by the C-suite, designed by crossfunctional teams and executed by line managers who are permitted to experiment and take calculated risks, companies can discover exciting new opportunities and take the swift and savvy decisions that deliver triumphs. 9
11 Appendix: survey results Percentages may not add to 00% owing to rounding or the ability of respondents to choose multiple responses. Based on your observations, how does your company s ability to gather and analyse big data compare to that of competitors within your industry? Industry leading (my company has a comprehensive strategy for gathering and gaining insight from a broad range of data sources) 7 Adequate (my company has a basic approach for gathering and gaining insight from a limited number of data sources) Somewhat inadequate (my company has a limited ability for gathering and gaining insight from data. Only a select few executives or specialists have this ability) Completely inadequate (my company has no formal way to gather and gain insight from more than a few data sources) 0 Don t know 40 To what extent do you agree with the following statements: Strongly agree Agree Neither agree nor disagree Disagree Strongly disagree Don t know Typically, businesses must empower more employees with access to big data so they can make informed business decisions 6 6 In general, the more data that are shared within an organisation, the more effective the decision making process The biggest barrier to getting useful insights and results from big data is the available technology In most cases, if data results are incomplete or inconclusive, it s the fault of the data source 7 The CIO drives all of our big data processes
12 Which job functions do you feel currently have the least extensive big data capabilities or limited developed data systems? Please select the top 4. Which job functions need to benefit from big data capabilities today? Please select the top 4. Customer service Human resources General management Strategy and business development Legal Marketing Finance 0 Risk/Security 0 Operations and production 9 Regulatory compliance 8 Sales 8 Information and research 6 R&D 4 Supply-chain management Procurement IT 0 Operations, real estate planning 8 Other 0 None of the above Customer service Marketing Strategy and business development General management Information and research Finance Sales Operations and production R&D 8 Risk/Security 6 Human resources Procurement Supply-chain management IT 0 Regulatory compliance 9 Legal 8 Operations, real estate planning Other 0 None of the above What kinds of business decisions in your organisation would benefit the most from the inclusion of big data? Please select all that apply. Market opportunity and segmentation Customer retention Product and service mix Competitor performance Financial performance Capital investments Risk mitigation Customer purchase Geographic market Fraud mitigation
13 Which of the following types of big data would your company benefit from the most in making strategic decisions? Please select all that apply. Which of the following data types would be most useful to analyze in combination with the above big data types in making strategic decisions? Please select all that apply. Internal data External website content Social media data Internet connected device data Location-based data Internal log files RFID data 9 Sensor-generated data 7 Other None of the above Competitive intelligence Customer segment data Market condition models and data Sales results and forecast data Internal customer data 9 Third party market data (eg, Demographics) 7 Point of sale data 0 Other None of the above Think about the last major decision your team made. Approximately what percentage of internal and external big data did you take into account? 0% 0% 0% 40% 0% 60% 70% 80% 90% 00% Now think about the result of that decision. What was the outcome? Positive Negative No effect (the decision had no impact) 70
14 Do you feel you had enough internal and external big data to make the decision? Yes No 4 If you did not feel you had enough internal and external big data to make the decision, to what extent do you agree with the following statement: Adding more sources of data would have helped ensure a positive business decision and outcome. Strongly agree 6 Agree Neither agree nor disagree Disagree Strongly disagree Not applicable (data did not play a part in the decision/outcome) 60 Over the next three years, do you expect your company to invest more in expanding big data decision capabilities to more employees? Yes No 47 What would you consider the biggest obstacles to expanding big data capabilities to more employees? Please select the top. Financial (high costs are prohibitive) Cultural (not an executive priority) Talent (skilled worker shortage) Security (too many variables) Technical (too complex to integrate) Organisational (not everyone would benefit) Logistical (too many data sources) Time (data capabilities take too long)
15 What do you consider to be the most important outcomes if your company expanded big data capabilities to more workers? Please select the top 4. Better decisions are being made faster Identifying business opportunities that were not previously apparent Identifying opportunities and exploiting them faster Planning scenarios with greater confidence Managing risks more effectively Increasing revenues 9 Bringing products and services to market faster 8 Adapting to change is easier Predicting outcomes with greater accuracy 8 Maximising business processes is easier 8 Reducing costs 0 Measuring outcomes has improved 0 Responding to third parties is more effective (suppliers, partners, customers, etc) 8 Other Don t know/cannot predict In which country are you personally located? In which region are you personally located? United States of America Canada 9 India 8 United Kingdom 6 Australia 4 Hong Kong, China, Singapore Italy, Netherlands, Belgium, Brazil, France, Greece, Russia Germany, New Zealand, Switzerland, Denmark, Hungary, Indonesia, Ireland, Malaysia, Mexico, Nigeria, Poland, Portugal, Romania, South Africa, Taiwan North America Asia-Pacific Western Europe Middle East and Africa 6 Latin America Eastern Europe
16 What are your organisation s global annual revenues in US dollars? Which of the following best describes your title? Board member CEO/President/Managing director CFO/Treasurer/Comptroller 0 CIO/CTO/Technology director Other C-level executive 9 SVP/VP/Director Head of Business Unit 8 Head of Department 7 Manager Other 7 4 $00m or less $00m to $bn $bn to $bn $bn to $0bn $0bn or more What is your primary industry? Professional services Financial services IT and technology Healthcare, pharmaceuticals and biotechnology Manufacturing Entertaining, media and publishing Consumer goods Energy and natural resources Education 4 Chemicals Automotive Government/Public sector Retailing Transportation, travel and tourism Telecoms Construction and real estate Agriculture and agribusiness Logistics and distribution What is your main functional role? 9 4 General management Strategy and business development Finance Marketing and sales 0 IT Operations and production R&D 4 Information and research 4 Supply-chain management Customer service Legal Risk/Security 7 8 0
17 Whilst every effort has been taken to verify the accuracy of this information, neither The Economist Intelligence Unit Ltd. nor the sponsor of this report can accept any responsibility or liability for reliance by any person on this white paper or any of the information, opinions or conclusions set out in the white paper. Cover: Shutterstock 6
18 London 6 Red Lion Square London WCR 4HQ United Kingdom Tel: (44.0) Fax: (44.0) New York 70 Third Avenue th Floor New York, NY 007 United States Tel: (.) Fax: (.) Hong Kong 600, Central Plaza 8 Harbour Road Wanchai Hong Kong Tel: (8) Fax: (8) Geneva Boulevard des Tranchées 6 06 Geneva Switzerland Tel: (4) Fax: (4)