Big data already paying off in insurance fraud detection

There’s a lot of talk right now about how big data is going to change insurance. Artificial intelligence and prescriptive analytics are definitely coming, and they will definitely change the insurance industry. To get a better idea of the specific ways in which big data applications are going to play out, it’s worth examining how data analytics are being used in the industry right now.

Many insurers’ initial forays into the world of big data focus on fraud detection. That makes sense — the Insurance Information Institute (III) estimates that 10 percent of property-casualty insurance industry losses each year are attributable to fraud, to the tune of $32 billion.

And most believe the problem is getting worse: 61 percent of property-casualty insurers report that the number of suspect frauds increased slightly or significantly over the past three years, according to a 2016 white paper issued by the Coalition Against Insurance Fraud. Insurance companies have always relied on technology to fight fraud. The III reports that 95 percent of insurers say they use antifraud technology, but about half say a lack of information technology resources prevents them from fully implementing it.

Enter big data. At the beginning of 2016, Towers Watson reported that 26 percent of insurance companies were using predictive analytics to address fraud potential. In the next two years, that number is expected to jump to 70 percent—more than any other big data application.

Putting text mining to good use

As data scientists team up with claims department leaders and other insurance professionals, many are looking to text mining as a crucial analytical tool for decoding enormous amounts of unstructured data.

Put simply, text mining is a way to scan large amounts of data for keywords, not unlike a web search. But claims departments are putting the technology to more sophisticated use, analyzing information for more significant data points and connections. Text mining can interpret claims adjusters’ handwritten notes and scan a claimant’s social media accounts for suspicious activity in nearly real time.

A shift in focus for claims

Text mining and other tools represent a fundamental shift in how claims departments seek out fraud. ACORD has stated that, thanks to big data, fraud detection will move from being claims-centric to person-centric. In other words, efforts to spot fraud will shift from focusing on the claim itself to the individual filing the claim.

Claims models will pull information about the would-be beneficiary from across claims, policies and external data sources, including information from other insurers, medical professionals, police, auto body shops and a host of other sources.

Privacy and accuracy concerns

This shift toward person-centric fraud detection and an increase in shared data raises issues regarding privacy and data quality. Individuals may opt out of sharing information with their insurers through vehicle telematics or social media, lessening the impact of insurer data analytics initiatives and creating a competitive advantage for less scrupulous organizations willing to harvest data against consumers’ wishes. Compounding the issue is the fact that data collected is not always accurate or easily manipulated.

Concerns about privacy and access to personal information are not exclusive to the insurance industry. However, insurance professionals on the front lines, from agents and brokers to customer service reps, can play a significant role in encouraging insurance consumers to share data by spelling out how the increased data benefits all parties. The benefits of additional data for reducing claims fraud, and the subsequent reduction in coverage costs, is a good place to start.

Ongoing need for talented insurance professionals

Issues surrounding privacy and data quality are still playing out at insurers nationwide. But as we’ve seen with other big data applications in the insurance world, talented professionals will still have a key role to play in the claims process.

Data mining techniques may be able to more readily identify red flags or spot patterns that suggest fraud, but insurers will still need to rely on data-minded teams to create the algorithms and systems that will detect and root out fraud. Building effective teams is not easy; it takes just the right combination of abilities and personalities, along with plenty of buy-in from the top.