I’ve been discussing anti-fraud programs for the insurance industry, and how analytics can be used to detect fraudulent claims. If you haven’t already, you may want to read earlier posts in this series:
- Part 1: Implementing an anti-fraud program.
- Part 2: Three lines of defense in an anti-fraud program.
Segmentation of claims
A side benefit of developing analytics for an anti-fraud program is the ability to segment claims. Using cluster analysis and pattern analysis, insurers can identify assignments and behaviors throughout the life cycle of the claim. These insights can benefit the claim-handling process to improve efficiency and effectiveness, and can ultimately help carriers achieve competitive differentiation.
Further, intelligence from claim handling can benefit the underwriting operation to create a better understanding of risk. This can translate into more accurate policy writing and pricing, which benefits both insurer and consumer.
Measurements of success
Once you have implemented an anti-fraud program, it is important to have metrics to determine whether the initiative has been successful. Possible criteria include:
- Reduction in rate of false positives. This will vary between carriers and depends on the sophistication of current fraud-detection techniques.
- Decreased cost of fraudulent investigations. As more sophisticated data analytics are used, investigations personnel have more data to draw from. This reduces the time required to investigate a suspect claim.
Anti-fraud programs benefit consumers
Though the challenges of implementing an anti-fraud program can be significant, the return on investment is quite high. Reducing the payouts due to fraudulent claims can result in significant savings for the insurer.
Equally important, as insurers shift focus to claims that are truly fraudulent, customer satisfaction and retention will improve. The cost savings from a successful anti-fraud program can be passed onto consumers, saving non-fraudulent customers money through lower premiums.