MIT Sloan Management Review PDF - What’s Your Data Worth?

What’s Your Data Worth? SPRING 2017 ISSUE Many businesses don’t yet know the answer to that question. But going forward, companies will need to develop greater expertise at valuing their data assets. James E....

What’s Your Data Worth? SPRING 2017 ISSUE Many businesses don’t yet know the answer to that question. But going forward, companies will need to develop greater expertise at valuing their data assets. James E. Short Steve Todd Vol. 58, No. 3 Reprint #58331 http://mitsmr.com/2mUheim [DATA & ANALYTICS ] What’s Your Data Worth? Many businesses don’t yet know the answer to that question. But going forward, companies will need to develop greater expertise at valuing their data assets. BY JAMES E. SHORT AND STEVE TODD I n 2016, Microsoft Corp. Caesars’ most valuable asset, acquired the online profes- its value to an outside party sional network LinkedIn was an open question. Corp. for $26.2 billion. Why As these examples illustrate, did Microsoft consider LinkedIn there is no formula for placing to be so valuable? And how a precise price tag on data. But much of the price paid was in both of these cases, there for LinkedIn’s user data — as were parties who believed the opposed to its other assets? data to be worth hundreds of Globally, LinkedIn had 433 millions of dollars. million registered users and approximately 100 million Exploring Data active users per month prior Valuation to the acquisition. Simple To research data valuation, arithmetic tells us that we conducted interviews and Microsoft paid about $260 collected secondary data on per monthly active user. information activities in 36 Did Microsoft pay a reason- companies and nonprofit or- able price for the LinkedIn user ganizations in North America data? Microsoft must have and Europe. Most had annual thought so — and LinkedIn revenues greater than $1 bil- agreed. But the deal generated deal between Microsoft and Journal article, “the most lion. They represented a wide scrutiny from the rating agency LinkedIn is just one recent valuable asset in the bitter range of industry sectors, Moody’s Investors Service Inc., example of data valuation bankruptcy feud at Caesars including retail, health care, which conducted a review of coming to the fore. Another Entertainment Corp.” A 2016 entertainment, manufactur- Microsoft’s credit rating after example occurred during report by a bankruptcy court ing, transportation, and the deal was announced. the Chapter 11 bankruptcy examiner on the case noted government. What can be learned from the proceedings of Caesars Enter- instances where sold-off Cae- Although our focus was Microsoft–LinkedIn transac- tainment Operating Corp. sars properties — having lost on data value, we found that tion about the valuation of user Inc., a subsidiary of the casino access to the customer analyt- most of the organizations in data? How can we determine if gaming company Caesars En- ics in the Total Rewards our study were focused in- Microsoft — or any acquirer — tertainment Corp. One area of database — suffered a decline stead on the challenges of paid a reasonable price? conflict was the data in Cae- in earnings. But the report storing, protecting, accessing, The answers to these ques- sars’ Total Rewards customer also observed that it might be and analyzing massive tions are not clear. But the loyalty program; some credi- difficult to sell the Total Re- amounts of data — efforts subject is growing increasingly tors argued that the Total wards system to incorporate it for which the information relevant as companies collect Rewards program data was into another company’s loy- technology (IT) function is and analyze ever more data. worth $1 billion, making it, alty program. Although the primarily responsible. While Indeed, the multibillion-dollar according to a Wall Street Total Rewards system was the IT functions were highly PLEASE NOTE THAT GRAY AREAS REFLECT ARTWORK THAT HAS BEEN INTENTIONALLY REMOVED. SPRING 2017 MIT SLOAN MANAGEMENT REVIEW 17 THE SUBSTANTIVE CONTENT OF THE ARTICLE APPEARS AS ORIGINALLY PUBLISHED. FRONTIERS What’s Your Data Worth? (Continued from page 17) effective in storing and pro- For example, one large finan- value. Here’s a breakdown of The frequency of data tecting data, they alone cannot cial group had a team working each value source: usage brings up an interesting make the key decisions that on a significant data classifica- aspect of data value. Conven- transform data into business tion effort that included the 1. Data as Strategic Asset tional, tangible assets value. Our study lens, there- categories “critical,” “impor- For most companies, monetiz- generally exhibit decreasing fore, quickly expanded to tant,” and “other.” Data was ing data assets means looking at returns to use. That is, they include chief financial and categorized as “other” when the value of customer data. This decrease in value the more marketing officers and, in the the value was judged to be is not a new concept; the idea of they are used. But data has the case of regulatory compliance, context-specific. The team’s monetizing customer data is as potential — not always, but legal officers. Because the goal was to classify hundreds old as grocery store loyalty often — to increase in value majority of the companies in of terabytes of data; after nine cards. Customer data can gener- the more it is used. That is, our study did not have formal months, they had worked ate monetary value directly data viewed as an asset can ex- data valuation practices, we through less than 20. (when the data is sold, traded, hibit increasing returns to use. adjusted our methodology to The difficulty that this or acquired) or indirectly (when For example, Google Inc.’s focus on significant business particular financial group en- a new product or service lever- Waze navigation and traffic events triggering the need countered is typical. Valuing aging customer data is created, application integrates real- for data valuation, such as data can be complex and highly but the data itself is not sold). time crowdsourced data from mergers and acquisitions, context-dependent. Value may Companies can also combine drivers, so the Waze mapping bankruptcy filings, or acquisi- be based on multiple attributes, publicly available and proprie- data becomes more valuable tions and sales of data assets. including usage type and fre- tary data to create unique data as more people use it. Rather than studying data quency, content, age, author, sets for sale or use. The major costs of data are value in the abstract, we history, reputation, creation How big is the market in its capture, storage, and looked at events that triggered cost, revenue potential, security opportunity for data monetiza- maintenance. The marginal the need for such valuation requirements, and legal im- tion? In a word: big. The costs of using it can be almost and that could be compared portance. Data value may Strategy& unit of PwC has negligible. An additional fac- across organizations. change over time in response estimated that, in the financial tor is time of use: The right All the companies we stud- to new priorities, litigation, or sector alone, the revenue from data at the right time — for ied were awash in data, and the regulations. These factors are commercializing data will grow example, transaction data col- volume of their stored data was all relevant and difficult to to $300 billion per year by 2018. lected during the Christmas growing on average by 40% per quantify. retail sales season — may be year. We expected this explosion 2. The Value of Data in Use of very high value. of data would place pressure on A Framework for Data use is typically defined Of course, usage-based management to know which Valuing Data by the application — such as definitions of value are two- data was most valuable. How- How, then, should companies a customer relationship man- sided; the value attached to ever, the majority of companies formalize data valuation prac- agement system or general each side of the activity is reported they had no formal tices? Based on our research, ledger — and frequency of unlikely to be the same. For data valuation policies in place. we define data value as the use. The frequency of use is example, for a traveler lost in A few identified classification composite of three sources of typically defined by the an unfamiliar city, mapping efforts that included value value: (1) the asset, or stock, application workload, the data sent to the traveler’s assessments. These efforts were value; (2) the activity value; transaction rate, and the cellphone may be of very time-consuming and complex. and (3) the expected, or future, frequency of data access. high value for one use, but the traveler may never need that exact data again. On the other hand, the data provider We define data value as the composite of three may keep the data for other purposes — and use it over sources of value: (1) the asset, or stock, value; (2) the and over again — for a very activity value; and (3) the expected, or future, value. long time. 18 MIT SLOAN MANAGEMENT REVIEW SPRING 2017 SLOANREVIEW.MIT.EDU 3. The Expected Future Value of Data Although the Making implicit data policies explicit, codified, phrases “digital assets” or “data and sharable across the company is a first step assets” are commonly used, in prioritizing data value. there is no generally accepted definition of how these assets should be counted on balance sheets. In fact, if data assets are A few companies in our need to address how to acquire information in data storage de- tracked and accounted for at sample were beginning to and develop valuation exper- vices and distributed systems. all — a big “if” — they are typ- manually classify selected data tise in their own organizations. Whichever approach ically commingled with other sets by value. In one case, the 3. Decide whether top- is taken, the first step is to intangible assets, such as trade- triggering event was an inter- down or bottom-up valuation identify the business and tech- marks, patents, copyrights, and nal security audit to assess processes are the most effec- nology events that trigger the goodwill. There are a number data risk. In another, the trig- tive within the company. In business’s need for valuation. A of approaches to valuing intan- gering event was a desire to the top-down approach to needs-based approach will help gible assets. For example, assess where in the organiza- valuing data, companies iden- senior management prioritize intangible assets can be valued tion the volume of data was tify their critical applications and drive valuation strategies, on the basis of observable growing rapidly and to exam- and assign a value to the data moving the company forward market-based transactions ine closely the costs and value used in those applications, in monetizing the current and involving similar assets; on the of that growth. whether they are a mainframe future value of its digital assets. income they produce or cash The strongest business case transaction system, a customer flow they generate through we found for data valuation relationship management sys- James E. Short is a lead scientist at the San Diego savings; or on the cost incurred was in the acquisition, sale, or tem, or a product development Supercomputer Center at to develop or replace them. divestiture of business units system. Key steps include the University of California, San Diego, in La Jolla, Califor- with significant data assets. defining the main system link- nia. Steve Todd is fellow and What Can We anticipate that in the fu- ages — that is, the systems that vice president of strategy Companies Do? ture, some of the evolving feed other systems — associat- and innovation at Dell EMC, a part of Dell Technologies. No matter which path a com- responsibilities of chief data ing the data accessed by all Comment on this article at pany chooses to embed data officers may include valuing linked systems, and measuring http://sloanreview.mit.edu/ valuation into company-wide company data for these pur- the data activity within the x/58331, or contact the authors at [email protected]. strategies, our research uncov- poses. But that role is too new linked systems. This approach ered three practical steps that for us to discern any aggregate has the benefit of prioritizing ACKNOWLEDGMENTS all companies can take. trends at this time. where internal partnerships The authors wish to acknowledge 1. Make valuation policies 2. Build in-house data val- between IT and business units financial and research support explicit and sharable across uation expertise. Our study need to be built, if they are not from Dell EMC, Intel Corp., and the company. It is critical to de- found that several companies already in place. Seagate Technology Inc. for this study; in addition, Cisco Systems velop company-wide policies in were exploring ways to mone- A second approach is to de- Inc., IBM Corp., and NetApp Inc. this area. For example, is your tize data assets for sale or fine data value heuristically — provided financial and research support for earlier stages of the company creating a data cata- licensing to third parties. in effect, working up from a research. Several individuals log so that all data assets are However, having data to sell is map of data usage across the made important contributions: known? Are you tracking the not the same thing as knowing core data sets in the company. Barry Rudolph of VelociData Inc., Douglas Laney at Gartner Inc., usage of data assets, much like how to sell it. Several of the Key steps in this approach in- Barbara Latulippe and Bill a company tracks the mileage companies relied on outside clude assessing data flows and Schmarzo at Dell EMC, and on the cars or trucks it owns? experts, rather than in-house linkages across data and applica- Terry Yoshii at Intel Corp. Making implicit data policies expertise, to value their data. tions, and producing a detailed Reprint 58331. explicit, codified, and sharable We anticipate this will change. analysis of data usage patterns. Copyright © Massachusetts across the company is a first Companies seeking to mone- Companies may already have Institute of Technology, 2017. step in prioritizing data value. tize their data assets will first much of the required All rights reserved. SLOANREVIEW.MIT.EDU SPRING 2017 MIT SLOAN MANAGEMENT REVIEW 19  MIT SL SLO OAN MANA MANAGEMEN GEMENT T REVIEW FRON FRONTIERS TIERS PDFs Reprints Permission to Copy Back Issues Articles published in MIT Sloan Management Review are copyrighted by the Massachusetts Institute of Technology unless otherwise specified at the end of an article. MIT Sloan Management Review articles, permissions, and back issues can be purchased on our Web site: sloanreview.mit.edu or you may order through our Business Service Center (9 a.m.-5 p.m. ET) at the phone numbers listed below. Paper reprints are available in quantities of 250 or more. To reproduce or transmit one or more MIT Sloan Management Review articles by electronic or mechanical means (including photocopying or archiving in any information storage or retrieval system) requires written permission. To request permission, use our Web site: sloanreview.mit.edu or E-mail: [email protected] Call (US and International):617-253-7170 Fax: 617-258-9739 Posting of full-text SMR articles on publicly accessible Internet sites is prohibited. To obtain permission to post articles on secure and/or password- protected intranet sites, e-mail your request to [email protected]. Copyright © Massachusetts Institute of Technology, 2017. All rights reserved. Reprint #58331 http://mitsmr.com/2mUheim Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.