As the insurance industry continues to embrace data analytics, the fraud sector is seeing increasing benefits of implementation both domestically and abroad.
Fraud is an international problem as business and manufacturing become more global. What impacts a company in Europe or Asia, will also affect companies in the U.S. and other countries. Cyber attacks such as the WannaCry virus are an excellent example of how one event can have global repercussions.
There are new tools, however, that will assist insurers and business owners in detecting and tracking the impact of fraud. Three insurance fraud professionals, one each from Germany, the U.S. and Canada, weigh in on how data analytics is changing the insurance fraud landscape. Michael Skiba, Ph.D., is vice president of counter fraud strategies for Inform and based in Germany. Jeffrey G. Rapattoni is co-chair of the insurance fraud/SIU practice group for U.S.-based Marshall Dennehey Warner Coleman & Goggin, and Chris McKibbin is a partner in the fraud investigation, recovery and enforcement practice group for Blaney McMurtry LLP, in Canada.
What are the efficiency gains for carriers using advanced analytics?
Skiba: Carriers that use analytics can maximize fraud detection and optimization with reduced efforts. One well-trained SIU analyst utilizing the power of analytics can filter and route hundreds (or even thousands) of claims; claims that would normally have to be reviewed using manual methods.
In this manner, the SIU units can focus on those claims that have the highest potential for success, making them more effective and efficient.
Rapattoni: Our clients are seeing a huge return by way of “time on task,” that is, the carriers are getting more done in less time. This gain allows a more detailed investigation on an inventory of claims that match a particular outlier. Because the process is now expedited, the initial spend on the fraudulent claim is significantly lower and the carrier is no longer chasing money.
McKibbin: Efficiency gains can be measured across multiple carriers as well. For example, the Insurance Bureau of Canada (IBC) now works with Canatics, a non-profit organization funded by nine member auto insurers representing some 75% of the market. Canatics pools data from its members and applies advanced analytics to identify suspicious activity. It then provides alerts to IBC, which investigates suspected fraudulent activity. The pooling of data from multiple carriers (not sharing, so as not to offend privacy regulations) has benefitted each individual carrier significantly.
How has the claims environment changed subsequent to analytic implementation?
Skiba: Companies that are using analytics in the claims environment are seeing significant return on investment. Many fraud solutions are, of course, used for fraud detection, but at the same time these systems can also be used in the claims environment. There are an infinite number of areas within the claim process that can benefit from the optimization process; whether it is fast-track settlement routing, workflow management, filling informational gaps with third party data, and many more areas.
Rapattoni: I think the claims environment is now faster. We are able to detect fraud quicker with technology than that of a manual read. It separates the claims inventory away from questionable claims so we can focus on process and payment to the insured.
McKibbin: Analytics has the potential to enhance all phases of the claims environment. Carriers are better-positioned to efficiently assign claims and set reserves; identify and fast-track low-risk claims; and detect potential fraud by using predictive analytics in real time.
What are the anticipated innovations/trends in the data analytics environment that insurers should be aware of for 2018?
Skiba: I think many carriers are aware that the future of fighting fraud is all about data; data management, inclusion and integration, and at the core of this data is the technology that surrounds it. But we can't forget the “people” aspect of what is to come in fraud fighting. Many companies are failing to recognize that they need people who have the ability to leverage the power of this data and, furthermore, have some degree of comfort using technology to assist with fraud fighting efforts.
This is not to imply that we will never need field-based SIU investigators “door knocking,” but we need to make sure all of our SIU staff is aware of the capabilities of technology and data in order to enrich their investigative skills.
Rapattoni: I would expect to see the role of the data analyst continue to expand. With the influx of technology in front of us, carriers will likely create roles for data experts so as to ensure their data platforms are being maximized for a positive return on investment. With that, I would also expect the number of fraud referrals to rise given the data scrub that an analytics platform will likely provide.
McKibbin: We anticipate that the positive lessons of analytics usage in the auto industry will more frequently be applied to other risks, such as P&C and life/disability insurance. The challenge will lie in reworking the relatively discrete set of variables in play in the auto sector for application in other market sectors.
Analytics are providing insurers with critical data to identify trends, opportunities and fraudulent claims. (Photo: Shutterstock)
How do the high-performing insurance carriers use data analytics to fight fraud?
Skiba: The high-performing carriers that use data analytics do so in a manner that is more holistic and broader than customary systems. These carriers will insert a fraud detection system, but also use that program in the point of sale, underwriting or claims environment. This will bring more data into the fraud system, data that will enrich and strengthen the fraud results.
Rapattoni: High-performing carriers can now identify fraud before the money is out the door. They have a more robust claims protocol that expedites the identification of fraud. With earlier detection, the carrier can choose to take a recorded statement or examination under oath or log the event for purposes of building a recovery action against the defrauding party. Simply stated, carriers are using this tool to stop the fraud bleed before it begins.
McKibbin: Michael's point about holistic approaches to data analytics is excellent. Underwriting and claims can each develop information useful to its counterpart for the ultimate benefit of both. It is essential that underwriting and claims consciously make an effort to avoid the “silo” approach, and realize that they can benefit each other immensely. The very best carriers will also engage in constructive self-criticism or “self-scouting,” critically assessing the outputs of their data analytics systems on an ongoing basis to improve what data is collected, how it is collected, and how it is processed and weighted.
How should global counter-fraud strategies utilizing data analytics differ from smaller scale “domestic” approaches?
Skiba: Carriers that have successful counter-fraud programs on a global level are ones that have realized the benefit of getting all of their systems “talking” to one another. It is vital for companies to refrain from a silo approach to their data; that is, not allowing information to leave certain areas. Carriers that can bridge the informational gap and analyze policy data, claims data and third-party data across global lines have seen incredible returns on this effort.
Rapattoni: Domestic approaches are more limited and oftentimes miss the big picture. The data is limited so the final product is limited. Carriers that take the global approach are always going to see the bigger picture because they are not focusing on limited sects. Taking a broader approach allows the carrier to consider fraud over multiple lines of business. Likewise, that approach allows the company to look beyond its own backyard and peer into different regions and countries. Bottom line; globalization of data detects more fraud.
McKibbin: The lessons of effective scalability play out in an even broader context in data pooling efforts across multiple carriers, such as that utilized by IBC and Canatics. Getting carriers in non-auto sectors to collect and record data with a view toward pooling of such data may represent the next frontier in this area.
Contact Michael Skiba at email@example.com, Jeffrey Rapattoni at firstname.lastname@example.org and Christopher McKibbin at email@example.com. Marshall Dennehey and Blaney McMurtry LLP are members of the Insurance Law Global Network (www.insurancelawglobal.com).