New technological developments are offering to resolve three issues in insurance - improving customer service, reducing the cost of claim handling and detecting more fraud. The solution is to use an application which integrates with existing claim handling systems and provides claim handlers with an instantaneous advise when they enter a new claim. This advice is based on experts’ insights into fraud combined with advanced analytics. It enables the claim handler to decide upon settling the claim immediately - thus offering excellent service and reducing costs -, performing a normal investigation, or sending the claim immediately to the fraud detection department in order to discover more fraud. To be more competitive and profitable, insurance companies are undertaking several initiatives to improve their claim handling processes, such as:

n Improving claim handling efficiency

The claim handling process is expensive, and makes up for around 20% of the total costs of an insurance company. Companies strive to decrease these costs by offering ‘fast track claim handling’. Companies want to handle 50-80% of their claims in a fast track, by being able to settle the claim during the first telephone conversation. This requires no expensive back office steps and no additional expensive customer contacts (in some cases, 60% of the phone calls into the claim handling center is from customers informing about the status of their claim). In addition, they aim to offer ‘customer self service’. Visionary companies want to further reduce the cost of claim handling by having insurants notify their own claims via internet, and handle these claims instantaneously, even without human intervention.

n Improving service levels to customers

The ability to offer customers a range of channels to notify their claims at their convenience and to respond quickly to these claim notifications enables companies to differentiate themselves and acquire and retain customers better than the competition.

n Decreasing the cost of fraud

Estimates are that up to 10% of all claims are fraudulent, however, insurance companies are able to discover only 0.5 to 1% of the fraudulent cases. In the past, insurance companies would simply accept fraud and increase premiums to make up for the increased costs of claims. Increased pressure from government and the public opinion is changing this trend now. The general opinion is that fraud is a crime, and it is unacceptable that other decent citizens should pay for these costs. Moreover, reducing the cost of fraud enables insurance companies to differentiate themselves by offering lower premiums.

Obviously, insurance companies are pulled between the need to reduce costs on the one hand, and improve customer service on the other. Technology, like real-time predictive analytics, can help insurance companies uncover two to three times as many instances of fraud, whilst enabling them to hand the first line of fraud detection over to the branch and call center teams who speak to customers every day. Such a strategy can vastly enhance fraud detection as well as enhance customer relations – and doesn’t require large investments in infrastructure or headcount.

Mind and data mining
Most fraud cases rely on gut feeling - many fraud experts explain that they can instantly spot a fraudulent claim: the time is actually spent substantiating their instincts. These instincts are based on considerable experience, but are often a handful of key risk indicators which can show how each case should be handled.

This knowledge is available within the fraud department, but not in other areas of the company. Institutions can gain considerable value by interviewing the experts and building their knowledge into the claim handling process, to automatically flag those claims that require further investigation. This “mind mining” process can help determine the handling of a great deal of claims, but it is not everything. Many fraud profiles are unknown to the experts but can be revealed by analyzing historic fraud data, to predict future trends: “predictive analysis”. Entirely new classes of fraud can be uncovered using predictive analytics technology – increasing annual fraud detection rates two or three-fold.

The fruits of both the mind and the data mining can then be developed into a range of “fraud profiles” – sets of criteria which establish whether new claims are fraudulent. If a claim matches any of these profiles, it can be sent to the fraud department for further investigation.

Closing the loop
The next stage is to put in place a process that helps to enhance fraud profiles on an ongoing basis – enabling companies to quickly understand new fraud trends as they develop. Predictive analytics is also highly effective here: fraud experts can dip into claims databases and use the technology to spot new trends. Thus fraud profiles are easily updated and fed back into the claim handling process. This is a simple, ongoing feedback loop of analysis and profiling by fraud experts, followed by deployment and action within the call center.

Insurers regard the speed they can respond to customers’ claims as an important differentiator. Some companies have found an innovative solution to this by setting up a “fast track” claims process, where 50% to 70% of all claims are handled within 48 hours. As well as the obvious customer service and sales benefits the strategy offers, this enables companies to significantly reduce their claims handling costs – by up to 40% in some cases. The success of this strategy relies on understanding which claims are unlikely to be fraudulent at an early stage. Without this understanding, the insurer leaves itself unacceptably exposed to fraud. The claim handling department needs to be empowered with the means to assess new claims.

The issue is that many claim handlers often lack the expertise to recognise fraud, and cannot match each claim to the stored fraud profiles. In a busy call center or branch environment it would be highly impractical to present them with a lengthy fraud questionnaire each time they register a new claim. To determine whether a claim is fraudulent and what action to take as a result whilst talking to the customer, they can be supported with real-time analytics: reducing the claims handling process to mere minutes with a handful of leading questions, and providing a real-time recommendation on whether the claim is potentially fraudulent and should be handled by the ‘normal’ claims handling process, or whether it can go though the fast track.

A one-stop claims handling process can open the door to a wide range of customer service, sales and marketing possibilities. Given careful investment in the right technology, and an emphasis on implementing the closed-loop processes outlined above, insurers can really catch up with fraudsters and deliver truly outstanding service to customers.