You've probably never heard of Signet Bank. A small, Richmond-based regional lender formed in 1987 by the merger of two even smaller banks, Signet was typically associated with the types of plain vanilla loans that make the executives at the financial behemoths in New York fall asleep in their chairs.
But that would soon change. In the early 90s, Signet made a bold bet: the company staked the future of its consumer lending business on two scarcely-known pioneers in the nascent field of applied data science whose ideas had been largely ignored by Signet's larger, global competitors in New York.
The two men, Richard Fairbank and Nigel Morris, theorized that they could use data related to the Signet's best, most profitable credit card customers -- even seemingly irrelevant information such as where they went to college, or how many kids they had -- to help them predict or model the profitability of new applicants, which would in turn allow the bank to offer more favorable terms to entice these higher-value customers.
There was one problem -- the data didn't yet exist. Signet, like all other credit lenders at the time, simply offered the same terms to all applicants, meaning that it had no way of modeling the outcome of offering different terms to different customers based upon its projected profitability. This means that the lender had no idea which variables would prove relevant in predicting this outcome.
The cost of acquiring this data would not be a cheap or easy investment -- to do this, Signet had to offer the more aggressive interest rates to applicants at random with little regard for their true value. Once the poster child for prudent credit decisions, the bank saw its charge-off rate double, costing the bank millions in the short term.
However, its investment would soon pay off. By 1994, the credit division had become so profitable that Signet spun it off into its own entity with Fairbank at the helm, renaming it "Capital One" -- now a $40 billion business. Perhaps you've heard of it: its competitors at the big banks in New York certainly have.
A similar revolution is underway in the marketing organizations at some of the world's largest consumer companies. In 2013, 85 percent of U.S. brand executives said that big data had yielded more than half of their marketing initiatives when it came to increasing insights into consumer behavior. Media buyers, who used to do business with suit-clad reps pitching magazine inserts, now take their vendor meetings with folks who dress more like Mark Zuckerberg and like to talk about the superiority of their 'algorithms'.
Forward-thinking retailers like Walmart and Target have notably embraced the big data revolution in merchandising decisions and couponing (sometimes too notably in the case of the latter), while CPG companies such as Kraft and Procter & Gamble employ teams of PhD-level data scientists whose work influences everything from supply chain decisions to diaper design.
When it comes to mobile advertising, however, even many of the most sophisticated marketers have a massive blind spot, similar to that of Signet's competitors prior to 1994. Despite a massive shift in eyeballs and consumer dollars to smartphones and tablets, very few marketers understand how to properly leverage the seemingly endless amount of data these devices offer, and as a result, have largely remained on the sidelines in applying data-driven techniques to their mobile advertising efforts.
In ecommerce, for instance, point-of-sale transactions on mobile devices have grown exponentially over the past two years and are now projected to account for 50 percent of sales by 2017. However, the vast majority of ecommerce companies still have not yet implemented systems capable of mining the data that's crucial to understanding how their costumers utilize different devices to research products and make purchases.
To understand the magnitude of this shortcoming, consider the comparatively mature space of desktop re-targeting. 90 percent of marketers now consider its efficacy to be on par with search and email -- traditionally the bread and butter, respectively, of ecommerce marketing -- and budgets have grown consistently and aggressively. For instance, platforms such as Criteo and AdRoll do a great job of finding users at their desktop computer and reminding them to buy that pair of shoes they were looking at earlier in the week (before inevitably being distracted by a cat video). However, their reliance on third-party cookies makes it difficult, if not impossible, to reach them when they're playing a game on their iPhone or watching a video on their iPad, which are cookie-less environments.
Even this example undercuts the power of mobile and the unprecedented opportunity it presents for marketers. Just as banks had access to reams of valuable consumer data that lay unmined before Signet came along, much of the most powerful data unique to mobile -- say, whether someone who has seen a car ad actually visits a dealership -- is never aggregated, analyzed, or utilized.
As desktop browsers' eyeball share continues to shrink and attention continues to fragment, marketers need to adapt by taking a device-agnostic approach. By implementing technologies like cross-device identification and dynamic creative optimization that leverage data science principles, they can make smart, real-time decisions to personalize the experience to each user and maximize ROI.
To be clear, many technological challenges still exist, and it is incumbent upon those of us on the vendor side to continue to work in tandem with our clients to find elegant solutions. But despite these shortcomings, marketers that take bold action and push the envelope will win, while laggards that sit on the sidelines will face increasingly stiff competition from their more progressive counterparts. As Wayne Gretzky, Richard Fairbank, or any good data scientist will tell you, you miss 100 percent of the shots you don't take.
On Twitter? Follow iMedia at @iMediaTweet.