Improved Fraud Detection Rates Using Dynamic Models For Insurers

Cut down on the number of fraudulent insurance claims without creating bottlenecks in your approvals

Medical claims fraud remains a major issue for insurance and medical service providers to track but is a critical area for the industry. Claims fraud costs the US healthcare industry billions of dollars a year, and it remains hard to prove, especially as the speed of claims and fulfillment accelerates with online insurance becoming more prominent.
The problem

An online insurance provider bills itself as one of the fastest to respond to user claims, but has recently run into a spat of fraudulent claims that are increasingly impacting its bottom line. The company has an established detection system that relies on using previous indicators to find similar abnormalities and red flags in new claims but has been unable to detect more sophisticated and subtle fraudulent medical claims.

The solution

Our customer needed a way to detect fraudulent claims earlier in the process but did not want to add more steps that could create bottlenecks and impact their user experience. The first step was to move away from a rigid rules-based detection system to a dynamic one. To reduce their time to fulfillment while building a smarter and more accurate detection system, the company focused on enriching its models with external data, including:

  • Economic information about financial stability, income, and insurance history
  • Social media interactions and history that could verify claimants’ identities and previous activity
  • Person data such including biographical information, internet usage, and search engine behavior
  • Alternative risk and credit scores that can point toward risk factors that would signify a greater chance of fraud
The results:

Improved accuracy and reduced bottlenecks

After logging in claim fulfillment periods that ranged as high as several hours, the company managed to reduce its average fulfillment time to roughly 1 hour per claim. More importantly, it managed to improve its detection rates by nearly 11%, boosting its overall detection rate to 92% of all fraudulent claims during the initial review. As a result, our customer was able to offer a much better service to their users, boosting their revenue and reducing churn. More importantly, they were able to do so while cutting down on fraud and saving significantly on each transaction.

Schedule a call with our team and learn how to migitage fraudulent insurance claims with better data.

Other resources you might be interested in

Case Studies: Melio Payments Case Study

Explorium Case Study - Melio is on a mission to keep small businesses in business by providing a smart, simple B2B payments solution tailor-made for their needs.

One-pagers: Explorium for Risk and Fraud

Fight risk with the data you need, not the data you have. Learn all about how Explorium supports risk and fraud use cases with external data and signals for better ML models.

Case Studies: Behalf Case Study

Behalf Uses Explorium-sourced Data to Accurately Segment Potential Business Customers, Improving Marketing Efficiency and Streamlining Operations

Get started with Explorium External Data Cloud Start for free