Improving Claims Fraud Detection in Insurance

Worldwide Industry Marketing Manager for Insurance, IBM

Solutions for analyzing big data can play a critical role in addressing the increasing prevalence of claims fraud. Traditionally, fraud is estimated to account for approximately 10 percent of insurance company losses[1], and that percentage is rising. Insurance companies need ways to quickly identify potential fraudulent claims, enhance the efficiency of investigations and prosecutions, and facilitate rapid reporting and visualization to improve ongoing antifraud efforts.

At the underwriting stage, insurance companies can employ solutions for big data that scrutinize applicant identities by searching and analyzing large volumes of information rapidly. Companies can determine whether applicants—and people associated with those applicants—have been linked to fraud in the past. Through a review process, companies can avoid fraud by denying applications for disability, health, homeowner or automobile policies with high fraud risks.

During claims intake, companies can use solutions designed to collect and analyze streaming data, such as social media posts or geospatial data, to inform investigations and policy decisions. This streaming data can help insurers discover, for example, whether policyholders are being honest about accident details or if services rendered are legitimate.

Predictive analytics solutions can help categorize risk and deliver fraud propensity scores to claim intake specialists in real time so they can adjust their line of questioning and route suspicious claims to investigators. For ongoing analysis of fraudulent claims and their impact on the business, companies can use solutions to analyze, report and create visualizations of data patterns.

The Insurance Bureau of Canada (IBC) is the national insurance industry association representing Canada’s home, car and business insurers. Because investigation of cases of suspected automobile insurance fraud often took several years, the company’s investigative services division wanted to accelerate its process.

The IBC worked with IBM on a project that explored new ways to increase the efficiency of fraud identification. The project showed how IBM solutions for big data can help identify suspect individuals and flag suspicious claims. IBM solutions also help users visualize relationships and linkages to increase the accuracy and speed of discovering potential fraud.

In the IBC project, more than 233,000 claims from six years were analyzed. The solution identified more than 2,000 suspected fraudulent claims with a value of CAD41 million. IBM and the IBC estimate that these solutions could save the Ontario automobile insurance industry approximately CAD200 million per year.

With enhanced information insight, companies can rapidly prevent, predict, detect and investigate potentially fraudulent claims and enhance the efficiency of ongoing antifraud efforts.

Learn more about the IBC project

Read the white paper, “Harnessing the Power of Big Data for Insurance

See more blog posts, videos, podcasts and reports on big data in insurance

[1]Insurance Fraud,” Insurance Information Institute, March 2013