Everywhere you turn, Big Data is touted as the solution, or at least the means to arrive at the solution, for so many issues and problems facing us today. It seems that, if applied properly, there’s nothing that a good heaping helping of Big Data can’t solve.

Now it’s time to add yet another item to the “Is there anything Big Data can’t do?” cheering section. Insurance fraud certainly doesn’t sound very sexy, but it’s responsible in part for nasty things like premium increases. People who say that insurance fraud is a victimless crime are full of it. We’re the ones who end up paying for those crimes, where it hits us right in the bank account. Anything that can help battle insurance fraud is more than welcome.


That’s an acronym for First Notice Of Loss. People file a FNOL to start the process of making a claim, basically informing their insurance company that they have sustained a loss that they intend to open up a claim for it. Typically, this used to be done by a phone call, but in the 21st century age of enhanced communication, FNOLs wind up also coming in by email, text, SMS, and even attached images.

With this barrage of information coming in, as well as subsequent follow-up reports and addenda, it’s easy for a carrier to get swamped, and therefore less able to spot inconsistencies, duplicate claims, or other less than proper practices. Thanks to Big Data, that sheer volume of information can be managed, and inaccuracies, both accidental and deliberate, can be caught.

Not So Fast, Naughty-spawn!

Big Data can head off a fraudulent claim simply by checking past customer information and comparing it to the current situation. Consider the following example. Someone files a claim citing the loss of personal property due to a house fire. But Big Data shows that similar claims were filed by other family members barely a year before, and a social media history shows that the claimant actually gave away several items to another family member a few months before. With the information gathered through Big Data, the fraud is caught before any claims get paid out.

The Cure For False Medical Claims

Here’s an area that’s ripe for abuse, especially when it comes to fraudulent disability claims. Say someone puts in a claim for a back problem, claiming that they were flat-out incapacitated and house-bound for eight weeks. And yet, other information culled from Big Data shows purchases made by the claimant at a sporting-goods store, reservations for two at a nearby ski lodge, and the purchase of lift tickets and toboggan rentals! Bringing all of this data together and spotting such discrepancies helps fake claims like this explode on the launch pad.

Less fraud in medical insurance means lower premiums and increased affordability, something which is good news for the 15% of Americans who don’t have health insurance, a figure cited by the article “Many Americans Still Lack Health Insurance”.

“Do You Want To Change Your BS Story, Sir?”

Or perhaps someone puts in a claim for an auto accident, citing loss of control of their vehicle in heavy traffic during bad weather, resulting in the inability to stop in time. But information pulled from Big Data sources show that not only was the weather clear, there was virtually no traffic on that roadway, and, most damning of all, according to state records, the car had failed inspection, particularly due to, yes, bad brakes. The only way that could possibly be worse is if the driver had a history of drinking … oh, wait, yes, according to his driving record, he’s received several DUIs in the last few years. So much for that accident claim.

What the above shows us is that Big Data, used in conjunction with analytics and data processing tools, can gather vast amounts of information from disparate sources, and assemble it into one very revealing, very complete picture. What was once a herculean task can now be easily accomplished, making it harder to perpetrate fraud.

So, whether it’s fighting fraud or helping you track down your lost dog, yes, it seems that there’s nothing that Big Data can’t do.