In a world of straight-through-processing and “touchless” claims, customers are demanding faster pay-outs on their insurance claims through smarter, more intuitive digital interactions and customer-centric support models.
Until fairly recently, standard industry processes have allowed time for loss adjustors, claims handlers and expert SIU investigators to appropriately assess customer claims. This approach has been a largely effective, but resource-intensive control on fraud. However, with the introduction of increasing automation, there is far less time available now for human review.
Insurers want to keep customers happy with smooth and rapid processes, but they also want to be confident that they are paying the right people, in the right circumstances, and limiting the opportunity for fraud. To achieve both of these objectives, real-time risk detection technology has a crucial role to play.
The far-reaching impacts of fraud
Insurance fraud has too often been regarded as a victimless crime. The reality is very different. Fraud has an immense impact on society, seriously damaging trust as well as creating material financial implications. According to the Coalition Against Insurance Fraud, criminals steal at least $80 billion every year from American consumers. Premiums rise to manage this additional risk, affecting all customers. Fraud also hurts loss ratios, disrupts daily operations, distorts pricing and affects reserves calculations.
And it’s not just insurance companies that pay when criminals carry out fraud—innocent people, often customers, get caught up in these crimes more often than most would like to think. Arson, murder-for-hire, crash-for-cash, staged accidents and medical malpractice are all examples where organized crime groups have targeted innocent citizens and exposed them to physical harm.
With insurance processes digitizing at an increased rate, the opportunity for fraud has expanded significantly, and it is vital that appropriate responses to those threats are available.
See also: It’s Time for Next Phase of Innovation
Demand more from technology
Anti-fraud technology has already evolved at an exceptional rate in the last five years, which has included the creation of better investigation tools and experimentation with data science or machine learning techniques But insurers should not accept yesterday’s technology when they can be pushing for tomorrow's:
- Scoring and alerting should be available in real time to keep pace with the demands from automated claims management workflows.
- Analytics should be transparent and explainable so that the work and decisions of investigators are defensible.
- Technology should be built on open architecture, should be capable of integration with core claims management or underwriting systems in real time and should offer flexibility for deployment in the cloud or on-premise.
- Expert investigators and skilled data scientists should be able to focus on the highest-value cases rather than being frustrated by mundane data tasks.
- Software and systems should support processes where appropriate, but insurers should be able to independently own and manage their own analytics without relying on external services.
Many companies find themselves working with siloed data, attempting to catch irregularities across unconnected data sets. Instead, insurers should demand a single view of all parties—policyholders, claimants, suppliers, brokers—to work within a single data set.
Insurers should also be able to use fraud management technology to easily detect and manage instances of identity manipulation—the slightest change between a name, date of birth, ID or address should be easily spotted and flagged, even if it’s across multiple data sets, to root out fraud without delay. Detection should also consider the relationships between parties, which is often as crucial to understand as the circumstances of each individual claim.
Data privacy regulation has changed significantly in the last decade with the introduction of new laws such as the California Consumer Privacy Act or GDPR in the EU. To ensure compliance. security models must be sufficiently granular and be able to support different user types in accessing different levels of data according to their specific permissions.
Finally, rather than opting for point detection solutions, analytics capabilities should be applied to deliver value across the enterprise. For example, intelligence that can be gained from a claims fraud detection solution can be highly valuable for detecting and preventing underwriting fraud. The same intelligence can equally be helpful in identifying churn risk or upselling opportunities. The technologies being deployed for fraud should also be sufficiently scalable and robust to service multiple use cases to maximize value and achieve a far lower overall cost of ownership.
See also: 7 ‘Laws of Zero’ Will Shape Future
Customers want to associate their insurers with stability, trust, competence and airtight operations. With so much innovation happening globally, now is the moment for insurers to think big and evolve their enterprise fraud capabilities. Fraud is not a victimless crime, it is not the cost of doing business and we do not have to accept the status quo.