Four ways digital can help combat insurance fraud
Fraud rings targeting the insurance industry are getting more sophisticated, but there is hope for carriers in the form of digital. That’s the message of a Strategy Meets Action research brief, “Fighting Fraud with Advanced Technology: Detection, Mitigation, and Prevention,” authored by SMA principal Karen Pauli.
There are four key ways that carriers can take action, the brief says:
- Leveraging predictive analytics
- Incorporating link analysis
- Working with new connected technologies
- Exploring artificial intelligence
Pauli writes that most insurers depend on business rules and workflow as the main technology backbone to combat fraud. However, that strategy is outdated, she adds.
“Even those insurers with a modern analytics environment face problems by relying exclusively on rules, models, and workflow,” the report says. “These tools must continue to evolve for continued effectiveness.”
The first step for insurers should be implementing predictive analytics in fraud investigation, Pauli writes. Just under half of insurers are already doing this, according to recent SMA research, but there’s no reason that number soon couldn’t be higher with a further 38% saying they plan to do so, she adds.
“This more nuanced approach is vital to counter the increasing sophistication of fraud rings,” Pauli writes. “For many, the timeframe for adoption needs to be accelerated.”
Another technology that can help combat fraud, in concert with predictive analytics, is link analysis. The technology helps companies detect correlations that indicate potential fraud rings. It’s common at larger insurers, but the technology is difficult to justify for small insurers, because of its cost. That’s where big data comes in, Pauli writes.
“For small and midsize insurers to make significant strides [in link analysis]… they may need access to consortium data to identify patterns that would be statistically insignificant within their own data,” she says.
Growing world of data
New digital data sources can pay off in fraud investigation, Pauli says. Chief among these is telematics, because of its wide range of business-line implications from auto to inland marine, but also its real-time, always-on nature. That provides a baseline of data against which suspicious activity can be measured.
“Accident reconstruction, injury assessment, force/location of impact can be correlated with severity of injury – and aggregating large quantities of similar data can identify unusual data points that may be cases of fraud,” Pauli says. Other connected technologies with promise in this area include wearable devices, the Internet of Things, and drones.
Finally, artificial intelligence also provides room for growth in fraud-fighting. Though the best solutions are probably still several years off, it’s a major focus area for the insurtech community, and insurers should prepare to integrate it in the future, Pauli concludes.
“AI is vastly more powerful than insurers’ existing analytics capabilities at detecting nuance and learning from experience – exactly what insurers need to combat organized fraud rings,” she writes. “Partnering with InsurTech firms may give insurers the accelerated adoption of AI that will result in a faster ROI.”