President Barack Obama’s credit card got declined in a New York restaurant. The likely reason, to quote Obama himself, was “…I guess I don’t use it enough, so they thought there was some fraud going on.”

While this is the most high profile example of predictive analytics going awry, it’s not the first and it certainly won’t be the last. Many of us have gone through the same experience, especially while traveling, making an outsized purchase, using the card too frequently - or too infrequently, as was the case with President Obama.

When a genuine cardholder’s transaction is declined, a “false positive” has occurred.

In a more generic scenario, the term refers to the condition in which the Predictive Analytics software raises an alarm which turns out to be false. False positives have been a major bugbear of Predictive Analytics technology ever since their inception. And, almost since then, we’ve been hearing that they’d decline as predictive analytics systems gathered more data, made more predictions and became more “self-learning”.

If someone believes that claim after the not-so-new technology just bungled with “the most powerful man on the planet”, well, I’ve the proverbial beachfront property to sell them in Arizona, London or New Delhi.

False positives is a major reason why, in my last blog post titled Difference Between Data Mining And Predictive Analytics, I'd rooted for diverting some of your analytics budget to Data Mining instead of using it all up on Predictive Analytics. 

As for Data Mining appearing to be unscientific, I suspect that's the result of following the purist approach of mining all kinds of data and letting the “chips fall where they fall”. 

Data Mining can get rid of that perception by restricting itself to areas that make business sense. When done right, it can deliver sensible and actionable insights, as it did in the following case (locales and other details changed):

“There’s a significantly higher attach rate of business loans with home loans in Zip Code 23508”

This finding makes business sense in hindsight when the bank discovers that its business banking and retail banking sales people sit in the same office in Norfolk, a practice that leads to better exchange of market information.

Data Mining experts could follow up their finding with the following recommendation:

“Let’s validate this finding in Norfolk. Then, let’s apply this practice to all our regions. If it works equally well everywhere else, our revenues will increase by 8% with no significant increase in cost or risk." 

With such a business-oriented approach towards mining data, Data Mining can help analytics overcome the risk of degenerating into hair-splitting and elevate the function to a strategic level.

While there are many implementation challenges around process, HR and so on, adopting Data Mining is a great way for analytics to gain a seat in the boardroom.