Can You Claim That Your Insurance Analysis Is Providing A Premium Result

Learn more about how insurance claims data can be analyzed to reduce fraud and prevent costly processing errors.

For today's blog post, I invited my colleague Qlik Director of Financial Services, Simon Kirby to give us his take on how analytics plays such a big role when it comes to insurance claims.

A few years ago I was working for the leading car insurer in the UK and we had a problem, a big problem. We were losing money to gangs of criminals that were deliberately creating accidents. They did this by braking hard so that the car behind would rear-end them. Insurance rules always meant that the car at the back was at fault because they were driving too close to the car in front. When the driver at fault was our policyholder, we (the insurance company) would have to pay the claim of the third-party driver which would often be thousands of pounds.

We had a suspicion that these third-parties were repeating these claims frequently and were never hurt as badly as they claimed. But the average compensation payments for “whiplash” claims in the UK were sufficiently high enough to encourage this behavior. A simple Google search shows how much claimants could potentially receive if their claim were paid:

Want to improve your detection rates of insurance claims fraud? #Analytics may be the answer:

I worked closely with my colleagues in the Claims Fraud team and we wanted to find repeat offenders that we could report to the police. However, we were struggling to analyze the vast quantity of data in our claims systems to find similar cases. Ideally, we wanted to find a collection of rear-end shunt claims with the same third-party driver, but this was proving to be a challenge. The third-parties often used a different name at the scene of the accident. Excel couldn’t cope with our data volumes and with our standard claims fraud systems, so it took weeks to change the search criteria and receive results.

So when I joined Qlik I wondered if the Qlik associative engine could help to solve this problem. The fact that Qlik Sense could compress the data and load all the fields into memory gave me the ability to analyze my data extremely quickly. I found that by combining the Qlik Sense bar chart with the feature called “Alternative Dimensions and Measures,” I could easily analyze my claims data and identify unique customer indicators such as the mobile phone number for the third-party. The interesting side-effect of this analysis was that I was also able to identify some minor data quality issues as well as some large and unusual claims transactions.

If you would like to see this in action take a look at this post on the Qlik Community website…
Here you can find the Insurance Claims App and a link to the associated video. Try it for yourself and let us know what you think!


In this article:

Keep up with the latest insights to drive the most value from your data.

Get ready to transform your entire business with data.

Follow Qlik