Technology to the Rescue in Fight Against Fraud
Insurance fraud comes in many shapes and sizes, from organized rings to opportunistic fraud, from parents fronting auto insurance for their teenage drivers to cyberfraud. The full scale of insurance fraud is unknown. Since the crime is designed to go undetected, the fraud-fighting community can only guess at the extent of crime and dollar losses.
While many policyholders still view insurance fraud as a victimless crime, it affects not only every insurance company, but also every policyholder due to increased premium rates. The FBI estimates that insurance fraud costs more than $40 billion per year or between $400 and $700 annually in extra premiums for every American family.
Historically insurance companies have seen fraud as a cost of doing business, but today carriers are turning to technology to combat this financial crime. A recent survey by the Coalition Against Insurance Fraud indicated that 95 percent of insurance companies are using anti-fraud technology.
Traditional fraud-detection systems using scorecards and profiling tend to focus on opportunistic claims fraud. These systems detect fraud at the individual customer or claim level, and overlook more organized criminal activity. But organized crime rings are growing, and so is the sophistication and velocity of their attacks. The anonymity of the Internet makes it easy for professional criminals to hide and shift identities and relationships, to evolve their tactics, and to disappear after a few successful transactions. The time is right for insurance companies to invest in technology to prevent claims fraud before it reaches epidemic proportions.
Successful insurers today must be proactive; putting more technology capabilities in place to prevent fraudulent incidents, detect fraud early, and aggressively manage fraud cases as they are identified. There is no one, perfect fraud-detection technique. Multiple techniques, working in concert, offer the best chance for detecting both opportunistic and organized fraud.
Insurance companies are using several emerging technologies to fight fraud:
In recent years, many insurers have turned to predictive modeling processes, using data-mining tools to build analytical models that produce fraud propensity scores. Predictive modeling tends to be more accurate than other fraud detection methods. Information can be collected and cross-referenced from a variety of data sources, resulting in more and better referrals than a traditional business rules system.
The claims process collects and generates large volumes of text-based information, such as adjuster notes, email, customer service calls, and police reports. In fact, unstructured data can represent up to 80 percent of claims data. Text mining software accesses the unstructured text, parses it to distill meaningful data, then analyzes the data to gain a deeper understanding of the claim. For example, you might use text mining to look for scripted comments in auto-accident claims. It would be a little suspicious if multiple claimants, allegedly unrelated, all say exactly the same thing. A new area of text mining is the ability to analyze the huge amount of data available within the social media world. Investigators are now searching Facebook, YouTube and other social media websites for possible telltale criminal evidence of the claimant.
Link analysis has proven effective in identifying organized fraud activities by modeling relationships between entities in claims. Entities may be defined as locations, medical network providers, telephone numbers, even attorneys, to name just a few. Large volumes of seemingly unrelated claims can be checked, and then patterns and problems identified. For example, link analysis might show a high-activity account with links from many accounts, or a low-activity account with strong links to a master account. It might reveal multiple claims in a short period of time from related parties, such as members of a single family, or the classic ring associated with staged accident scams.
Insurers have successfully used link analysis to identify the presence of organized fraud rings and take appropriate action. Furthermore, using these linking and network scoring techniques, not only can insurers avoid paying fraudulent claims at first notification of loss, but they can also check new policies for connections to historical fraud to avoid proliferation of fraud.
Big data analytics
Data is the most valuable commodity for any anti-fraud technology. The current focus on big data analytics is driving innovations in fighting fraud, especially in the ability to process large and complex data sets very quickly. One of the most exciting aspects of big data analytics is the ability to use data sources that were previously ignored because they were often too large or changed too frequently for more traditional analytical approaches. Big data analytics represents an opportunity to completely revolutionize the way fraud is detected.
Fraudsters are highly adaptive and continually change tactics, strategies, and even modes of operation. To combat fraud, insurance companies need to invest in anti-fraud technology to become more effective and evolve their strategies as fraud schemes shift. This was confirmed by the survey from the Coalition Against Insurance Fraud that found over one-quarter of insurers reported an increase in IT budgets for anti-fraud technology, primarily for link analysis and predictive modeling.
An anti-fraud strategy that includes the right mix of tools and technologies will result in a much higher fraud detection rate. This strategy will go a long way toward cutting overall losses for an insurer, which will translate into more accurate pricing, a competitive edge and lower premiums for policyholders.
(Stuart Rose is global insurance marketing director at SAS, a business analytics software provider. Connect with him on Twitter @stuartdrose or via email at Stuart.Rose@sas.com)