(Image credit: Shutterstock.)
Customer expectations for faster, more convenient service have been key drivers of innovation for insurers. But another consumer trend is also affecting change—lax attitudes about insurance fraud.
Consumers have been showing increasing tolerance for fraud in recent years. This shift comes at an interesting juncture—fraud is increasing, and insurers are paying claims quicker and easier thanks to digital acceleration. In fact, some insurers are pushing toward straight-through claims processing.
But convenience and improved customer experience can’t be at the expense of a strategic defense—especially considering the rising cost of fraud. Recently, the Coalition Against Insurance Fraud reported that fraud costs the industry a staggering $309 billion annually. That figure, along with digital trends, makes it more critical than ever to invest in advanced anti-fraud technology.
Fraud compounds operational challenges
Fraud comes in all forms and across lines of business, and, unfortunately, it’s increasing. Organized fraud schemes continue to rise, and opportunistic fraud tends to spike during difficult economic times, making the current high-inflation environment fertile ground for fraudsters.
Meanwhile, carriers face operational challenges such as determining which claims to streamline and which to investigate; managing cycle times, exposure, and litigation; and operating in leaner environments with stretched resources. On the micro level, carriers’ special investigative units experience their own challenges with the lack of quality referrals, the limited number of cases they can accept and the effectiveness of investigations. These issues put significant pressure on claims organizations, which only increase as resources decline, claims automation expands and fraud rises.
The advantages of fraud risk scoring
Risk scoring technology offers tremendous promise in addressing the fraud issue. The right scoring tools can help quickly identify potential fraud risks and automate detection across claims, involved parties, and medical and service providers.
To understand how risk scoring works, it may help to visualize a huge funnel that consumes all an insurer’s claims and scores them for risk in real time and from first notice of loss. Only the claims with the highest propensity for fraud are slowed down for further investigation while the lower scoring claims can be routed for straight-through processing to lower overhead costs and reduce cycle times.
Today’s risk-scoring tools also incorporate advanced analytics such as AI and predictive analytics to not only score claims but also explain why a claim or party scored as it did. Predictive modeling, for example, examines variables in historical data to predict future outcomes. It employs a more proactive approach to fraud detection. Currently, approximately 80 percent of recently surveyed large insurers use predictive modeling in fraud detection, and it is one of the primary areas of future investments for carriers overall. But there are challenges to operationalizing predictive models and realizing a return on investment.
Bigger data, better results
Data is essential to anti-fraud technology—both the quality and quantity of data. But accurate data can be elusive. According to a survey, 64 percent of insurers said poor data quality is the main challenge to implementing anti-fraud technology.
Data enrichment can help solve the data quality issue. It transforms prepossessed data into a comprehensive profile that can support data analysis and risk scoring for in-depth insights. There are several interesting data sets available to help generate quality analytics, including: claims history data, bankruptcy and foreclosure records, civil and criminal records, weather reports, vehicle location images, and prior loss images. These records can provide tremendous insight into entities, businesses and involved parties to help improve the accuracy and effectiveness of predictive models.
The amount of data matters greatly as well. The more data available for a predictive model, the better it will be at improving scoring accuracy and identifying questionable claims. That’s why it’s important to consider how a fraud scoring solution combines or aggregates data.
When multiple carriers across the industry add their data together, the data set becomes much larger and more powerful. Without aggregated data, a carrier is limited to its own historical data and potentially some small third-party data sets. And that can be dangerous, as fraudsters often perpetrate schemes across multiple insurers to avoid detection.
It’s more effective to recognize suspicious claims with a broad set of data across the industry, lines of business, coverage types, and involved parties. Aggregated data can also be used as a basis for network analytics to uncover hidden connections and fraud rings, and medical provider fraud analytics to identify suspicious billing practices.
Anti-fraud tech is the core of claims automation
Digital transformation in the insurance industry continues to accelerate, and anti-fraud technology needs to be at the core of that innovation wave. Aspirations to further automate the claims-handling process are closely tied to having advanced fraud analytics in place.
As insurers explore emerging anti-fraud tech, there are several questions they should consider, such as whether the solution integrates with their claim systems, if it automates referrals, can they customize risk scoring based on their book of business, and, of course, what is the potential return on investment? These critical questions deserve thoughtful answers as insurers journey through the ongoing digital transformation and aim to strengthen their perimeter defense against ever-increasing fraud.