Consumer Data: It’s Not Just What You Have, It’s Also Knowing What to Do With It.
SEE LAST PAGE OF THIS REPORT Paul Sagawa /Artur Pylak
FOR IMPORTANT DISCLOSURES 203.901.1633 /.1634
psagawa@ / email@example.com
October 2, 2014
Consumer Data: It’s Not Just What You Have, It’s Also Knowing What to Do With It.
As the time that consumers spend on line continues to grow, particularly on mobile devices, the information that various companies are able to collect about their users is also growing. The value of that information – for commercial advertising, commerce, and personalized services – allows those companies to better address markets that are collectively worth $17T. We have built a taxonomy of that information – demographics, interests, actions, locations and connections – and assessed the degrees to which each category might be valuable. We then created a framework to assess various companies with stakes in the flow of consumer data based on their access to data, their ability to collect and analyze the information, and their potential for monetizing it. Not surprisingly, GOOG is far ahead in all regards – the extensive data they are privy to across the taxonomy, their prodigious institutional skills in data collection and analysis, and the levers for monetization that they are able to pull. Amongst the rest – AMZN, FB and TWTR are obvious leaders. AAPL, and NFLX are constrained by their ability/willingness to monetize. Other internet players – YHOO, AOL, YELP, EBAY, etc. – lack scale and depth in the data that they are able to collect and reach in the vehicles used to monetize it. The same is even more true of the incumbents in industries in the path of digital competition, including banks, retailers, media companies, and others.
- Ubiquitous mobile computing yields a trove of data on individual consumers. The growing use of cloud-based services wherever and whenever a consumer goes throughout their day affords companies involved the opportunity to collect data. We have categorized that data into 5 buckets: Demographics, Interests, Actions, Locations and Connections. In each case, the collected data can be assessed by the breadth of the attributes collected, the length of time that the data covers, and the quality (accuracy, sharpness, etc.) of the data itself. It is also important to consider the degree to which the breadth of the data is integrated into unique profiles directly tied to specific individuals vs. generalized by segment.
- Consumer data key to online attack on markets worth up to $17T. Detailed data profiles allow online companies to target specific, high likelihood consumers in merchandizing and to tailor services to maximize their value for individuals, advantages that expand dramatically with the reach, breadth, quality and time span of the data collected. Currently, digital advertising ($140B worldwide), e-tail ($1.5T), and digital content ($57B), are the biggest businesses constructed on this data, but there is significant room for future growth in each of these categories and new opportunities to exploit data, including personal financial services, health care/fitness, home/energy management, transportation, and others.
- We evaluated 20+ large cap companies for their potential to exploit consumer data. Many companies have access to information about consumers stemming from mobile platforms, media distribution, search/navigation tools, financial transactions, and communications. We assessed leaders from both online and traditional business models for their access to valuable data, their ability to collect and analyze that data, and their potential for monetizing it. This exercise reveals a world of haves and have nots, where the few companies able to draw unusually high quality data from frequent engagements with a large population of attractive consumers over a sustained period reign supreme.
- GOOG, FB and TWTR are winning the ad game. GOOG has a hammerlock on search data, the gold standard for flagging interests and purchase intentions. Add in the Android platform funneling users to GOOG’s oft used, well integrated, and ad friendly applications under a single sign-in, a peerless data analysis infrastructure and a hand on every monetization lever, and GOOG takes the overall prize. FB has an enormous horde of data on its 1.2B fanatically engaged users, but the data is a step below in quality and monetization is uneven vs. GOOG. TWTR has a stronger signal for its users’ interests and clearer context for advertisers than FB, but a much smaller universe of less engaged users.
- AMZN, AAPL and NFLX have narrow data strategies. AMZN captures demographic, interest and purchase data from its 162M MAUs, using it to recommend products and build engagement/loyalty, key drivers for its $70B e-tail franchise. It is exploring monetizing through advertising, but remains a relative neophyte. AAPL, with a loyal base of 200M iPhone users and 800M cards on file, has considerable reach but is unprepared to collect or exploit data about its customers. NFLX uses a deep understanding of its customers’ viewing preferences to acquire content and guide recommendations, but eschews advertising and has a very narrow set of data that is tied to accounts rather than individuals. Other internet players – e.g. YHOO, EBAY, PCLN, YELP, AOL, etc. – lack data on multiple dimensions, and will be disadvantaged in competing in their core businesses and in entering new ones vs. the data leaders.
- Traditional players will be severely challenged. Media, retail, and financial services players, amongst others, face significant long-term challenges from data rich online competitors. Typically, traditional companies have limited data sets, lack the skills to collect and analyze that data, and are inexperienced in monetizing it. TV networks and distributers hope to use digital ad insertion to target consumers, but progress is slow, online video viewership is rising, video ads on FB, TWTR, etc. have gained traction, and NFLX could eventually include an ad driven alternative. Retailers are behind in tracking and targeting consumers, overmatched in their IT capabilities, working with data limited to their own transactions and often, tied to credit card numbers which will be obsolete once tokenization is fully implemented. Banks and card nets lust after advertising, but operate under strict regulations, may be overestimating the quality of their data and their abilities to analyze and monetize it, and are at longer term risk of disintermediation as mobile payments brands (a la Apple Pay) slip in front of them with consumers and merchants.
- Privacy little obstacle for consumer data leaders. While sharp criticism of top consumer data driven companies’ data policies has raised concerns amongst consumers, it has not hampered engagement with these services and very few users choose to opt out of data tracking when given the chance. We believe that the long term value of consumer data is huge, and that the economies of scale, scope and skill are broadly separating a few winners from many losers. We see GOOG, FB, AMZN and TWTR as obvious winners, with AAPL and NFLX, perhaps shortsightedly, standing on the sidelines despite strong potential data assets. We are sanguine about other online consumer data plays, e.g. YHOO, AOL, YELP, etc., given the severe scale disadvantages. Most traditional players will struggle as consumer data becomes increasingly valuable to their core businesses and online competition becomes ever more powerful.
Knowledge is Power
In a world of ubiquitous mobile connectivity, some companies know a lot about you. Because you log in – to your phone OS, to your email, for online shopping or social networking – GOOG, FB, AMZN, TWTR, AAPL and others can keep detailed records on you, not just with their own apps, but from other sites where those credentials are used to establish identity as well. This information comes in five different flavors: 1. Demographics – Your name, address, age, family status, social security number, telephone number, email address, etc.; 2. Interests – The things that you really like, dislike and want to know/do, including what you want to buy; 3. Actions – What you are doing (sites visited, apps used, media viewed, items purchased, etc.); 4. Location – Exactly where are you?; and 5. Connections – Who are your friends, family members, business associates, and group affiliations? This data varies by its breadth, its duration, and its quality, aspects that vary considerably amongst those companies that collect it.
We evaluated the consumer data positioning for 26 companies, both online players and threatened incumbents, assessing their access to data across the five categories, their ability to collect and analyze it, and their potential for monetizing it. Amongst digital ad driven companies, FB, TWTR and, particularly, GOOG stand out, for the breadth of data to which they have access, the size of their user bases and in their abilities to analyze and monetize. The gap vs. other online advertising players – e.g. YHOO, AOL, YELP, etc. – is considerable and growing. Of the companies choosing to exploit consumer data through direct e-commerce, AMZN is the obvious winner and has an eye toward pulling the advertising lever as well. Other US commerce sites – e.g. EBAY, PCLN, etc. – lack scale, engagement, and breadth. Two other possibly strong data players remain on the sideline. AAPL, with tight control of the iOS platform and its 200M iPhone users, apparently collects little of the available user data, evidenced by the woeful record of its iAd network and its merchant pledges around Apple Pay. NFLX uses detailed analysis of its users browsing and viewing to recommend content and acquire attractive programming, but thus far, has eschewed advertising. We suspect that NFLX will eventually revisit this stance.
The shifts toward digital advertising, streaming media, integrated merchandizing, and online financial services are fueled by the savvy use of this consumer data and a real threat to incumbent players. Media companies, retail merchants, banks, and other traditional consumer facing businesses largely lack access to data on many dimensions, and, typically, do not have the skills or infrastructure to exploit the data they may have. Cable operators brim with optimism for digital ad insertion, many top retailers are forging ahead with plans for their own digital wallet for payments and customer loyalty, and banks and credit card nets look to move their awkward dance of co-opetition into digital leadership based on transaction data. With rare exception, big talk on consumer data by these players has amounted to very little actual progress.
Meanwhile, it’s full steam ahead for GOOG, FB, AMZN and TWTR. The privacy issues that raise activist and governmental hackles have had no apparent impact on user growth, engagement or revenues. We believe that data targeted digital advertising can more than double its 15% share of the nearly $1T global ad market. We believe that superior merchandizing will aid data advantaged online retail players to sustain better than 20% annual growth through the end of the decade. We believe that data-fueled programming decisions will help NFLX (and others?) capture significantly more time from more viewers. We believe that a superior understanding of consumer needs and interests will allow digital platform owners opportunity to integrate payments and other services to their benefit.
Exh 1: Average Daily Consumption of Media
I Always Feel Like Somebody’s Watching Me
On the average day, the typical American consumer spends 5 hours and 46 minutes online, checks his/her smartphone more than 125 times, searches the internet about 7 times and shares roughly 0.3 photos. About 6.5% of all US retail transactions now take place over the Internet. All of these numbers are growing at an ample double digit pace YoY, with little sign of slowing (Exhibit 1).
A fact of this increasingly digital life is that all of this online activity is recorded, stored, and tied to a digital profile based on your log in to your mobile platform, your social network, or your email. Many companies then use powerful computing infrastructures to analyze the data across categories and over time to gain insight into your preferences and likely behavior. To the extent that these computer models prove to be accurate predictors, the companies then use the insights to target advertising, recommend products, and to improve the performance of the services that they provide for you. The more comprehensive the data is, the better the targeting, the recommending and the improvements in performance will be. The more consumers upon whom a company has a digital profile, the more valuable the data base will be in generating opportunities for generating revenue.
This is playing out in many corners of the US economy. Digital ad platforms are pitching agencies and advertisers, competing with TV, radio and print, based on superior targeting and tracking of consumers. e-tailers are reaching out to customers with special offers, based on past purchases. Mobile devices are reminding you to leave for appointments based on predicted commute times, offering restaurant tips in new cities, and recommending morning news items, based on what time it is, where you are, what you like and what you’ve done. Streaming platforms use your history to serve new music and recommend new videos that you are likely to enjoy. This personalization increases the usefulness of online services for consumers, while simultaneously improving the efficacy of messages sent their way and the efficiency of digital transactions.
Exh 2: SSR’s Taxonomy of Consumer Data
Who, What, Where, When, How and Why
To better understand the data being generated by online activity, we have defined a taxonomy of the various bits of information that are collected. Separating digital consumer information into five categories – demographics, interests, actions, location and connections – we can assess the value of the data for the various monetization levers in use (Exhibit 2). That value will vary based on the size of the population, the breadth of the data available, the time span for which the data has been collected, and the quality (i.e. accuracy, resolution, completeness) of the information. It will also vary based on the degree to which it has been integrated into a single profile tied to a specific individual consumer. Taking the categories one by one:
Demographics – Most demographic information about users is captured during a registration process used to establish log in credentials for an app or website. Typically consumers volunteer their names and email addresses, identifiers that are often accompanied by more personal data items, such as telephone numbers, personal stats (age, sex, relationship status, kids, etc.), home addresses and profile photos. Commerce sites will usually have credit card numbers, the answers to “personal” questions for authentication, and shipping information for various addresses. Financial institutions may have even more detailed data, including social security numbers, bank accounts, personal assets and liabilities, credit scores, etc. Led by Apple, consumers are increasingly asked to contribute biometric identifiers, such as fingerprints, heart beat records, or facial scans, as well. Demographic data is critical to establishing a consistent profile to which to assign all of the other data that is collected, and for managing authentication of identity. It can also be an important qualifier and/or identifier to assess the applicability of a message or service to an individual.
Interests – Many apps purport to have detailed graphs of their users’ interests, but the breadth and quality of those profiles varies widely. For example: a consumer putting an order for a specific model of an appliance in an e-commerce shopping cart is a strong signal of interest in that product, while another user posting a photograph of herself standing near that same appliance in a social networking site is a weak signal. In the first instance, the user is at the edge of the decision, and while he may ultimately abandon the shopping cart, there is an excellent chance that the user can be influenced into completing the sale. In the second instance, the juxtaposition of the consumer and the product may well be random, or even a sarcastic joke. Amongst commonly used apps, shopping gives the strongest indication of interest, followed by general search, while social apps largely infer interest based on “likes” and “follows” that may be based more on personal relationships and projecting a desired image than on one’s actual preferences.
Actions – A platform – like the OS on a mobile device or a browser on a PC – can learn a lot about you. It knows what apps you have used, what websites you have visited and what media you have consumed. It knows how much time you have spent playing games or watching streaming video. It knows what appointments you have made and details for many of the tickets that you have bought. It may also know about purchases that you have made, particularly if confirmations are sent to the email address attached to the platform profile. App companies will usually know only those activities that a user undertakes while in their specific app, but some, in particular Facebook, Google and Twitter, have recruited other apps/sites that will use a common log in for identification. This saves users the time of registering for a new app and the trouble of remembering a long list of passwords, while supplying the provider of the social log in an additional source of data on user activity. For actual transaction activity, the information is currently captured by multiple participants. The merchant knows who you are, what you bought and where you are having it shipped, often able to connect this with your in store purchases via a loyalty card or your credit card number. N.B. The move to tokenization by credit card issuers will destroy the value of the card number for identifying customers, neutering some merchants’ customer tracking efforts. The bank knows everything that you have purchased with that particular card, tied to a rock solid ID profile. A mobile or online payments app also knows what you have bought and where, although Apple Pay is looking to set a precedent by NOT collecting this data for its own use.
Location – A mobile device platform, like Android, iOS or Windows, is able to track a user to almost any location as they use their device, creating a record of each day’s movement. From this, the platform can infer addresses for commonly visited spots, presuming that a consistent final stop is home and a location visited each weekday is either work or school. The platform may also check your calendar and email to assess where you are likely to be in the future as well, anticipating vacations and work trips by appointments and travel reservations. Apps will know some of this as well, logging where you are each time you access the app (or log in affiliated apps) and noting when travel plans are mentioned in posts or if transportation or hospitality reservations are made through e-commerce. As they adopt low power Bluetooth technology, like Apple’s iBeacon, merchants will be able to sense customers who have enabled apps on their devices as they enter their stores and communicate directly with them. Knowing where you are, or where you are going, creates an opportunity to anticipate your needs and to offer valuable geographic context to target ads or tailor services.
Connections – Social networks deal in the currency of connections. Your associations – friends, family, acquaintances, colleagues, professional peers, favorite businesses, and group affiliations, etc. – reveal important inferences about your interests and provide important signals for presenting content. These connections are gathered by apps and by platforms via email and telephone contacts, friend lists, likes, and follows. These connections are also an important avenue for distribution, fueling viral content and propagating an app’s user base. We note that connections are associated with significant network effects that greatly advantage businesses able to establish critical mass – users have major disincentive to shift away to an alternative service once these connections have been established.
Exh 3: Selected Markets Benefitting from Consumer Data
Knowledge is Money
Information has proved to be a substantial asset for digital firms. Most of Google’s $65B in annual revenues accrue from its understanding of those things for which consumers are looking. Digital advertising, with superior targeting and tracking as its calling cards, is already a $140B global market, with designs on the remaining $800B or so still spent on traditional advertising categories. Retail e-commerce is a $1.4T global business, driven primarily by cost and convenience but advantaged by the savvy use of customer profiles to promote loyalty, accelerate sales conversions, promote new products, and reduce transaction friction. Despite rapid growth, online is still well less than 10% of total worldwide retail sales, leaving substantial further runway to continue the upward trajectory (Exhibit 3).
Longer term, broader swaths of the economy will be vulnerable to information advantaged digital competition. Payments is currently in the spotlight after Apple Pay’s bold entrance – although Apple has pledged NOT to collect and exploit consumer data, the rivals that come after will make no such pledge and all of these services clearly erode the brand value for both banks and card nets for a digital future. With the erosion of the relationship of the consumer to traditional payments brands, digital platforms will be positioned to push further into personal financial services – currently a $500B+ world-wide market. Certainly financial institutions have considerable customer data of their own, but they are highly regulated in the use of that information and have shown fairly limited expertise in using it for anything other than security and credit analysis. Online players, with similar data collected from unregulated activities, will look for space to get in and will use data to do it.
Other areas could also be addressed by information-rich digital services. Health care, although wrapped by seemingly impenetrable layers of regulation and bureaucracy, would seem an obvious candidate for incursion by data driven innovation. Apple and Google have both added interesting health/fitness data collection and analysis capabilities to their mobile platforms, along with sensor packed wearable accessories. While we believe it will take several years for all of this to really affect the way health care is delivered, the potential is enormous. Transportation is another area bound by regulation and tradition but ripe for disintermediation. Uber has clearly struck a blow to the status quo, perhaps leading the way for innovations like Google’s autonomous vehicles or Amazon’s drone delivery service. All of this will be informed by customer data. In summary, advertising and e-commerce alone make a $16T near term addressable market, but a decade out, information-based players may be looking at a market that could be even double that already massive number.
Exh 4: Companies Evaluated Against Consumer Data Taxonomy
The Wheat and the Chaff
We evaluated 26 large cap companies from both the digital world and traditional markets, based on their access to consumer information, their ability to collect, manage and analyze it, and finally, their levers for monetizing it (Exhibit 4). We applied the Consumer Data Taxonomy qualitatively to each of these companies and also quantified all three levers into scores ranging from 1-5, with 5 the best. The result is a composite score with a maximum of 15 for companies that are best in class for using consumer information. We believe that all three elements are critical to the value that the various companies can generate from customer data. For example, many traditional businesses have theoretical access to information that they are not able to collect or analyze due to inadequate systems infrastructure and a lack of institutional data analysis skills. Presuming that a retailer will be able to implement a cutting edge customer loyalty and digital merchandizing strategy just because the technology exists for them to begin collecting transaction data from their stores is an overbold assumption. Similarly, there are digital players that have access to valuable data and the capability to analyze it, but that choose not to monetize it as fully as possible. Netflix, which eschews advertising and Apple, which denies interest in transaction data that is clearly available to it, are the obvious examples. Finally, there are also companies that have strong data analysis capabilities and a clear interest in monetization that, unfortunately, do not have access to a particularly strong data set. Many subscale internet businesses – YHOO, AOL, YELP, etc. – come to mind. Commenting on several of the most important players:
Google – Arguably, Google was the first company to really make consumer information a digital business, and its search franchise remains the best current model for collecting and monetizing data profiles (Exhibit 5). Through Search, Chrome, the Android platform, its social sign on program, and its archipelago of well integrated apps – Maps, Gmail, the Play store, Plus, Now, Wallet, Shopping, YouTube, and others – Google tracks its users across devices, gaining critical data across all five categories, all of it tied back to a secure and authenticated log in profile. Within this, search data has proven the most valuable, as clear indicator to advertisers of interest and intent. Critics of the company have expressed concern that search activity on mobile devices is being subsumed by focused apps, but we believe that much of this phenomenon involves replacing searches for specific sites that were not easily monetized on the desktop either.
Google logs over a billion unique visitors to its web sites each month, enjoys data from over nearly a billion Android mobile devices, provides search functions for more than 2 billion devices, both mobile and desktop. More than a billion people access YouTube each month and 500M users use Gmail. That is a lot of uses and a great deal of engagement across the many linked apps, earning Google a top score in reach. The breadth of its information is also top notch, with a strong play for all five categories aided greatly by its market leading Android platform. Given its early play on information, it also gets an obvious top score in the length of its data set, while leadership in search and other reference resources for users give it another one in the quality of its data. Google’s ability to collect, manage and analyze data is FAR ahead of all other companies, not just in the consumer space but overall, and it has proved deft in pulling monetization levers, with a dominant digital advertising business, a growing e-commerce presence and interests in a variety of other potential sources of revenue.
Facebook – Facebook is far and away the leader in connection data, with its well honed social graph on the more than 1.4B registered visitors that its app and website receive each month (Exhibit 6). Facebook is also the leader in social sign in, allowing it to capture data as its users visit a variety of other sites and apps as well. Because its users are also heavily engaged, spending 4.7 hours a week, making 19.4 posts and uploading 2.3 photos on average, it also has a good record of activity and can log it to wherever the user was when accessing the app. However, the pattern of clicks, likes and favorites that Facebook uses to assess its users interests may not be the most accurate way of establishing interests – many are the “likes” used to express support for a friend rather than an actual affinity – and the inference of possible buying interest less than direct, dinging it a hair vs. Google in the Quality of its consumer data. Still, Facebook gets high marks across most of the categories, showing strong breadth and duration against leading marks for reach.
Facebook has a very strong data analysis capability, second only to Google and, perhaps, Amazon. It has also been very successful in monetizing its data via advertising, showing better than 80% YoY growth in ad sales in recent quarters as a clear number two player to Google. We believe that Facebook’s monetization success will continue, as it breaks its monolithic site into more activity focused apps in the mobile context, giving users a cleaner experience and advertisers a much better context for advertising. We also believe that Facebook can be a successful platform for the distribution of advertising supported short form video, creating another potential monetization lever.
Amazon – Amazon’s platform plays – the Kindle Fire and the poorly received Fire Phone – are a relatively small universe of users for the company (Exhibit 7). Most of its consumer data comes from its successful online retail business and, in particular, its popular Prime program which ties free shipping and streaming video for an annual fee. Amazon’s reach is considerably less than either Google or Facebook, with roughly 250M active customer accounts and a universe of Prime customers which is believed to be a bit over 20M worldwide. Still, Amazon’s information on its customers is of very high quality – actual purchases, product browsing activity and products lingering in the digital shopping cart are the strongest possible indicators of future purchasing interest. The information is integrated back to a very complete profile, which includes shipping addresses and approved payment accounts, to remove friction from the transaction. Of course the breadth of Amazon’s interest and action data is limited to the activities, mostly shopping but some media consumption and product review posts, that take place on its sites, and its connection data is related to wish lists and gift shipping for friends and family.
Amazon’s ability to analyze its data is right there with Facebook, and a step behind Google. It uses the data to recommend products, to tailor its services and increasingly, to direct advertising. In this, the company has obviously pulled hard on the e-commerce monetization lever and appears very open to pursuing other ways to exploit the information that it has on hand. Amazon rates third on the list primarily because of its limited reach and the relatively narrow range of its data.
Twitter – Like Amazon, Twitter suffers from a relatively short reach – it has 271M monthly average users, along with an unspecified number of visitors who view tweets without registering for the service and who generate considerably less value (Exhibit 8). The engagement of those users is good, not great, but the signals that they provide to their interests, based on who they follow and what posts they click, are very strong. Twitter’s view of actions is limited to activity on the site, which consists primarily of choosing tweeters to follow, browsing tweets, and occasionally, clicking through to consume media content. Location data is seemingly unimportant to Twitter’s monetization strategy, while connections data is asymmetric and more reflective of interest that actual personal connection.66
Overall, Twitter’s reach is mediocre, the breadth of its data is narrow, the timeframe is adequate, but its quality is high. Twitter has also shown excellent ability to analyze and exploit its data, another step behind Amazon and Facebook, but well ahead of most everyone else. This is reflected in a strong trajectory of monetization – ad sales have been up more than 125% YoY in recent quarters. Twitter has also shown a flair for innovation in its monetization, introducing new rich media ad formats successfully, negotiating strong partnerships with traditional media, and recently introducing an e-commerce angle with one-click to purchase advertising.
LinkedIn – LinkedIn has a focused use case around professional connections that gives it access to unique data and monetization opportunities (Exhibit 9). The company has 313M registered users and a reasonable track record of engagement. It can boast an interest graph around work-related topics and can tie it to unusual demographic data around its users’ employment histories and skill sets. It also has a strong picture of the professional networks connecting its users.
Overall, LinkedIn’s data set ranks similarly to Twitter’s, with decent reach, fairly narrow breadth, reasonable timeframes and very high quality, due largely to its focus and rarity. LinkedIn’s ability to analyze its data sets is good, and it has exploited its data to create a lucrative recruiting and job search service based on that data that will be difficult for others to replicate. It also monetizes via advertising, a revenue stream that accounts for about 20% of its sales and is growing at a strong 46% YoY pace.
Netflix – Netflix knows a lot about its 50M subscriber households. It knows what programs each household browses and what each watches (Exhibit 10). It knows if an account watches a program to the end or if drops it mid-stream or fast forwards to the end. It knows when someone goes back and watches a program a second time, or subsequently watches a program featuring the same actor or directed by the same person. Netflix uses that information to recommend programs to its subscribers, increasing the value of the service to those viewers by helping them find content that they will enjoy. It also uses the data to guide its own purchases of content, both originals and from 3rd parties. While Netflix has a very interesting data set on a fairly narrow, but growing, base of users, it has been unwilling to move beyond the monetization lever of its own subscription sales. We believe that there is considerable potential for the company to do more, particularly in offering advertising driven alternatives at lower price points.
Apple – Apple is another company that has spurned opportunities to monetize information about its customers (Exhibit 11). As a platform owner with unprecedented control over the activities in which its users engage, Apple could, conceivably, have detailed profiles of all of its users across all of the categories in the taxonomy. However, Apple’s efforts to expand into broader, data driven initiates have been failures – iAd and Ping have been notable – to the point that the company now seems to want to promote its lack of data as a benefit to consumers. During the recent launch of its Apple Pay initiative, management was explicit in touting its hands-off approach to data.
So Apple gets high marks for the data to which it could conceivably have access. It does have more than 200M iPhone users and 800M iTunes accounts. However, Apple denies interest in collecting or analyzing the data created by those users, except in a few incidences, such as recommending music and video. While Apple professes disinterest in collecting customer data, it also has limited institutional skills for managing and analyzing it, a perfect hat trick for low scores in the mechanics of the data business. As such, it has also been weak at monetizing information assets, preferring to simply take a toll as 3rd party apps, like Google, Facebook, Amazon and Twitter score revenues from iOS devices.
Microsoft – Microsoft has some interesting consumer franchises (Exhibit 12). It has 400M Outlook accounts, 56M Bing users and more than 50M active Xbox live accounts. The consumer PC population may be shrinking, but there are still nearly 100M in use in the US alone. Microsoft also has very strong data analytical capabilities – right up there with Amazon and Facebook, behind Google. Still, Microsoft has been disjointed in its attempts to monetize its consumer data, owing, perhaps, to a lack of integration across its data sources, to the relative weakness of its mobile platform, and a lack of direct engagement with consumers through its apps. Microsoft’s new CEO Satya Nadella has given little indication of an interest in consumer data as an opportunity, and it may be that the focus on the enterprise market leaves the advertising and e-commerce monetization opportunities relatively unexploited.
Other Digital Players – With a few exceptions most other online businesses are too small with too little reach to gain much value from the customer information that they may be able to collect (Exhibits 13-14). We are setting aside the big Asian messaging and e-commerce players due the peculiarities of the geographic market that they serve. Yahoo could conceivable make value from the 450M MAUs that they receive to their suite of web services. Still, shrinking engagement, uninspired advertising numbers and poor performance as a social log in speak to real weakness in the breadth and quality of data that Yahoo is able to capture and exploit. The same is true in spades for AOL and others with a general consumer bent. More focused internet players, like Yelp, Priceline and others, are quite vulnerable to the larger platform players with longer reach and broader data, at least in their access to information driven monetization.
Traditional Media – TV networks and cable operators have great hopes that they can stem the tide of digital media and steal some of its data driven mojo by using a technique called digital insertion, which will allow specific ads to be targeted at specific households (Exhibits 15-16). While questions remain as to the timeframe with which this capability will be deployed to the nation’s cable systems, we are also skeptical as to the quality of targeting data being captured by cable operators and the ability to effectively exploit it. Cable operators know what channels each cable box are tuned to, and can tie it back to the demographic information provided in the account record. We are skeptical as to the quality of that demographic information – customer records searchable by fixed line telephone numbers and street addresses will be hard to integrate with other targeting information. We are also skeptical as to the ability of either system operators or networks to parse that data into targeting signals for advertisers that go much beyond the neighborhood and typical viewer data that is used for general TV ad placement today.
Traditional Retailers – Retailers have become very serious about customer data (Exhibits 17-18). The MCX consortium of 70 national merchants, led by WalMart and including Target, Dunkin Donuts, and many others, is building its own mobile payments e-wallet solution with two goals in mind – reducing the transaction fees charged by the financial industry and reserving the data about those transaction for their own exclusive use. That second goal speaks to ambition. Retailers have seen Amazon use online transaction data to build its enviable Prime customer loyalty program and to drive sales through effective promotion, and they believe that they can do the same. We see this as a mixed bag. Very few individual retailers will have the reach of Amazon. The world’s largest retail operation, WalMart, boasts 200M customer visits to its 11,000 stores each week, certainly with some repeat visits week to week, and has 61M unique online visitors to its site each month, but the statistics for the rest of the retail world are certainly much, much less impressive.
Today, much of the consumer data captured by traditional retailers is tied to credit card numbers and will be rendered invalid as the card industry rolls out tokenization which blocks the actual card number from the merchant at POS. Even those companies with well established loyalty programs typically capture only a small percentage of sales against those profiles, as users have grown a bit weary of the myriad of cards they must carry to participate in the programs. Moreover, knowing the sales history of a shopper in your store is a small fraction of their total shopping behavior, and could be misleading in predicting future interests. The bigger and broader the merchant, the better will be their data in breadth and quality. The ability to actually analyze and exploit the data will also differ wildly from merchant to merchant. It is safe to say that no traditional retailer has data analytic ability comparable to Amazon or Google, and we believe most of them will not be up to the task of building a useful data driven loyalty program without a lot of help.
Financial Players – Facing off against the merchants are the card issuing banks, which feed on the fees charged to each transaction made. Banks have a trove of deeply valuable information – transaction histories across merchants, credit histories, asset assessments, reward program activity, etc. – but are highly regulated in their ability to use it for purposes outside of the financial services that they provide directly to their consumer customers (Exhibit 19-22). While they possess substantial data analysis prowess in the service of credit analysis and fraud prevention, it is not clear that the expertise can be easily applied toward predicting consumer behavior for advertisers or merchants. Card networks also have access to transaction data in their role as intermediary between the banks and merchants, but have not traditionally collated that data against the individual account holder demographic profiles. While the card nets clearly aspire to a stronger role in exploiting this transaction data, it is not clear that they have either the regulatory freedom or the analytical skill sets necessary to succeed. For both banks and card nets, the recent move by Apple to bury the financial brands behind Apple Pay is a bit ominous with regard to their future ability to exploit transaction data for advertising or marketing services.
Industry activists have made privacy a cause célèbre, particularly in Europe. Nonetheless, consumers have voted with their online activity – engagement on Google and Facebook, the primary targets of privacy activists have hardly suffered for the publicity around privacy issues. By their behavior, consumers do not share the concerns of the activist class. While regulations, such as the recent requirement by EU regulators for Google to erase references to specific individuals on request, have layered new responsibilities on Google, the end result has been de minimus. Users have not punished online sites for their liberal use of personal data to drive advertising and commerce – in fact, the popularity of Google and Twitter has only expanded in the midst of Privacy activism. While government intervention, particularly in Europe, where privacy issues have more currency, is a distinct threat, we remain confident that intervention will not dramatically upend the trends toward information use.
We believe that the savvy use of user data will be critical to success in advertising and selling to consumers going forward. To that end, several companies have significant advantage in collecting and exploiting information about consumer demographics, interests, actions, location and connections, based on their access to data through their existing core businesses, and their institutional ability to analyze and monetize it. We see consumer information and the ability to effectively exploit it as a major asset for several cloud-based companies, with Google the obvious leader. Behind it, we see Facebook, Amazon, Twitter and LinkedIn, in that order, as significant beneficiaries of the information driven consumer Internet.
Netflix and Apple are significant for their resolute refusal to fully exploit the data to which they have access, although we are more hopeful that Netflix will move to monetize its information assets than we are Apple. Other online players lack either sufficient scale or the ability to analyze and monetize their data. Similarly, we see serious risks for smaller online players, who we see as having insufficient scale and institutional skills to exploit the consumer data to which they have access. We also believe that incumbent players in media,
TV distribution, retail and consumer financial services are overestimating their ability to profit from their own customer data, particularly as data advantaged rivals have targeted their core businesses