Her, one of the current holiday movies, recounts the story of a man, one among many in a period about 10-20 years in the future, who falls in love with his phones’ operating system (OS). The OS, with Scarlett Johansson as its voice, has been built incorporating what she says is the DNA and knowledge of some of the best software developers in the world. By asking a few diagnostic questions of the user, the OS customizes its style and becomes a friend.
Falling in love with an OS is not so far fetched in a near-future world in which the film’s protagonist works for a dot com that provides handwritten letters for people incapable of expressing themselves, whether to the husband they have been married to for 50 years or their child. The recent Martin Scorsese American Express commercial in which he chats with Siri as if she is a personal assistant also suggests the possibility of a more meaningful relationship.
While the movie never uses the phrase, “Big Data” allows this love to occur and makes it almost inevitable. Big data allows the OS to understand its user, based upon analyzing similar user types; it allows the OS to customize itself so that it can be unique to the male user (while also being “unique” to many others); it allows the OS to anticipate the user’s desires and needs, perhaps even before the user himself knows what he wants.
This reality, a bit unnerving in the context of the film, already exists within banking, with financial services providers commonly applying it to consumer activities in order to develop product offers and suggest product to buy. However, big data’s capabilities are at least equally applicable to commercial banking and, increasingly, leading banks, whether large or small, use the information and direction that this type of analysis can provide to increase productivity, capture more wallet share, and better target product sales activities by relying on propensity models that predict with a high degree of accuracy the most likely next product that a target will buy.
As described in the appropriately named book, Big Data, the writers state that the advent of greater reliance on big data depends on three mind-set shifts: the “ability to analyze vast amounts of data;” a “willingness to embrace data’s real-world messiness;” and “a growing respect for correlations rather than a continuing quest for elusive causality.” Let’s look at each of these three elements.
* Vast amounts of data. There is an unprecedented amount of data available today. Example: our data partner has access to over 35 million risk-rated, geo-coded businesses in the U.S. and Canada and can apply over 200 proprietary and non-proprietary data bases in developing its analysis. It also possesses wealth and wallet share information on over 20 million owners and executives.
* Messiness. The authors write, “Big data transforms figures into something more probabilistic than precise.” Modeling using big data allows bankers to more effectively target customers and prospects. That said, 90%+ confidence means that 10% of the time the product recommended for sale may not be the right one. As one manager who uses this approach stated, “I tell the sales staff that it is a great tool. If it sends us in a wrong direction one time, just go on to the next name. ” Management has to operate with a tolerance for imperfection, not really much to ask, is it? In addition, as firms like ours gain more data from clients and other sources, its quality improves even further. One of the tough elements for many of us to get our heads around is the concept that “more [data] trumps better [and more selective data].” Much evidence shows that it does.
* Correlation over causality. For me, this concept is both the most intriguing and, on some level, the most terrifying. Consultants spend a lot of time analyzing situations and developing analyses and recommendations based on causality; for example, consultants often begin to design a project by developing hypotheses that the project will either prove or alter. Big data replaces a hypothesis-driven approach with a data-driven one. As the authors write, “Predictions based on correlations lie at the heart of big data…the correlations show what not why … knowing what is often good enough.” In other words correlation-based analytics may tell a bank with a high degree of statistical certainty that a company, based upon its past purchases, is likely to purchase product X. That involves a very different process than asking a relationship manager to conduct a yearly account review and set product priorities based upon his knowledge of the client and the product set. In effect, in the big data world we do not care why a certain product is the likely next purchase, we just know that it most likely is.
Ultimately, big data will significantly impact and change the role and responsibilities of relationship managers and business bankers. RMs, many of whom have viewed themselves as artists who should be able to design their own jobs based on their interests and capabilities, will find their jobs redesigned for them with resulting higher performance goals.
Once banks accept the value of big data in the commercial space, how can management best apply it? Examples include:
- Highlighting and targeting businesses currently operating with consumer DDAs, allowing for cross sales and repricing.
- Identifying consumers who are also business owners with accounts at other banks, increasing the ability to gain household wallet share.
- Quantifying business wallet share opportunities, increasing per account product sales.
- Selecting priorities for marketing based upon client-specified criteria, enhancing productivity.
- Directing calling activity on a branch specific basis, providing top priority customer and prospect names, setting clear directions for branch staff and establishing accountability.
- Providing lending and wealth management opportunities aimed at business owners, further locking them into the bank.
- Uncovering specific customers and prospects that “look-a-like” the bank’s most profitable customers, creating clear priorities for limited sales staff.
- Recommending changes to marketing strategies, products, sales practices, organization, and other areas impeding success, addressing the critical issues that constrain growth.
When I mentioned some of these capabilities to a banker from one of the top five banks, he reacted by saying, “I thought all banks already could do this.” Over the past decade or more his bank had the insight and the budget to invest millions in this area, creating a proprietary information base. But, today, the benefits of big data are readily available to the $1B bank as well as the $1T player.
While the future for banks is not quite as good as having Scarlet Johansson (or the male equivalent) whispering in one’s ear, big data provides great leverage and actionable knowledge, as close to a “secret sauce” as we are likely to get. For more information on this topic, consider attending our upcoming webinar, next week on January 14th: