There is a new weapon in the fight against fraud - cognitive interviewing. Bill Jackson explains
Would you like to be able to detect fraudulent claims before they even get beyond front-end staff? Philip Swift of Claims Management & Adjusting says cognitive interviewing can help you do just that.
Swift and his colleagues went to VFM Services to see how it might adapt its cognitive interviewing course to the motor claims business. VFM says the techniques have helped to detect discrepancies and inaccuracies in a range of claims.
Legal & General has also said the method has led to a 20% reduction in its household claim costs. So why have insurers been so slow to adopt this method?
Cognitive interviewing was developed as a process for screening for fraud in refund claims for lost or stolen travellers cheques made by customers calling from all over the world. The results, now well documented, were beyond anyone's expectations
When insurers conduct an investigation into a claim it is seldom with a criminal prosecution in mind. Yet, still the industry perseveres with traditional, criminal-style fraud investigation techniques. It is evident that at the initial claims handling stage, fraud detection and risk assessment skills are in short supply. Here, there is often an acceptance that there is no answer to fraud, or that salvation lies in technology.
The human element
But, no matter how clever or powerful software is at identifying risk, at some point someone somewhere has to make a justifiable and defensible decision - a role that can only be played by a human operator with the right skills. Cognitive interviewing can be adapted and applied to a range of investigative situations.
It was, in fact, not intended for use on any type of insurance claim. Originally the pundits said insurance was `different' and the process could not be applied to insurance claims. And even now, after Legal & General has used a derivative of the system to reduce household claims costs by more than 20%, there is still a reluctance to accept there is a way to manage fraud that works.
The question is, how well equipped are claims handlers to deal with the risk? It seems many have not received training in how to ask questions in a way that will enable them to understand fully what has happened or to test a fraudster's account.
Listening properly
In addition, they wrongly perceive fraud control and customer care to be exclusive activities. Having asked a claimant a question, they do not listen properly to the answer. When looking for risk, they focus on the personal attributes of the claimant rather than on the act of the deception.
They also rarely accurately capture the detail provided by a claimant at any stage of the claims handling process. So when the claim is identified as being suspicious, valuable evidence contained in an earlier conversation is no longer available.
Providing front-end staff with the skill to identify risk will not only lay a sound foundation for any ensuing investigation, it will also `encourage' a large number of `opportunist' fraudsters to retreat from the process at an early stage. Genuine customers making a claim will not only be unaware their claim has been assessed for risk, but will also be delighted at the speed with which their claim has been processed.
Training will not only reduce fraud, it will minimise leakage, increase claims handlers' morale and, because the process encourages professionalism in claims handling under a structured process, will result in additional savings in handling costs.
Bill Jackson is managing director of VFM Services (01462 457 910)
The rationale behind cognitive interviewing
Source: VFM Services
Case study: applying cognitive interviewing to suspect claims
by Philip Swift of Claims Management & Adjusting
Insurers are faced with identifying three types of insured; the genuine customer, the opportunist who exaggerates and the fraudster, whether opportunist or professional.
We knew that genuine customers must be identified early on in the process and we wanted to provide the fraudster with a means by which to withdraw the claim early in the claims process.
Cognitive interviewing seemed a good way of achieving these aims. It is nothing new, common sense suggested that if it was applied appropriately in the claims environment, our efficiency could be improved. Cognitive interviewing requires the interviewer to possess a wide insurance knowledge, patience, and attention to detail and good communication skills.
We entered into discussion with VFM services, which has a reputation for training personnel in such skills and, even though the system was not originally designed for motor claims, we thought the methodology could be applied with development on our part.
VFM personnel visited our offices prior to the course and spent a day familiarising themselves with our procedures. This resulted in the course being adapted for vehicle-orientated workload. Capturing the loss circumstances was no different to the process associated with a household or travel claim. Where we identified differences and a need to customise the process this was undertaken.
The training environment included practical work, interspersed with theory. It was not long before we were applying our newfound skills and practices to tailored examples.
The course was not rocket science, but principally about managing conversations and identifying deceptive behaviour so that we could make confident decisions. Questioning is not as straightforward as it sounds.
Unlike in a face-to-face interview, where one side of the conversation is written down longhand, the questions are recorded, enabling both sides of the conversation to be considered later. By recording the call, as opposed to constructing a written statement, a natural flow of conversation is developed.
It is important that we do not tell an insured what we think happened. They are encouraged to speak; the genuine claimant will gush, the fraudster struggles.
Contrary to popular belief, vehicle claims are complex. They require a specialist skill. Documents and dates need to be painstakingly compared; the devil is in the detail. Incorporating these elements while remaining true to the training did not prove as difficult as at first anticipated. The result: a process by which we can take over a claim at any stage and confidently progress it to settlement.
We returned from the course keen to implement the processes.
Within days an insurer was faced with a dilemma; its insured appeared to be making excuses to avoid an appointment, but offered to provide anything required over the phone. The insured was pressing for settlement in respect of a motorcycle, stolen from his garage.
Without anything more than a faxed claims form, the interview was conducted. The claim for £4,500 was exaggerated, a realistic settlement was £3,800; the insured would accept this. However, when questioned about the loss circumstances the insured faltered; there was no garage. Why was this misrepresented? The insured believed that the claim would not be entertained, suspecting his insurers would not provide cover at his inner-city address unless he had a garage. He was also aware that this premium was likely to be higher.
In the weeks we have been applying the course knowledge, the most notable claim was that of a £36,000 Mercedes.
The vehicle had been left outside a friend's home while the insured stopped by to collect some belongings. Thirty minutes later the insured returned to where he had parked the car to discover the theft. According to the insured, keys had not been used to facilitate the theft. The insured said that all sets of keys had been declared to the insurance company.
In assessing the risk, the handler first asked the insured to provide an account. A second account was pursued, with detailed notes taken; the insured's replies to controlled questions were added to a time line. There was a lack of reasonably expected detail, missing steps and evasive responses. These were more noticeable when the keys were mentioned.
Probing the responses left us in no doubt keys were used to take the car. Further enquiries confirmed our findings.
We feel the facts may have been uncovered by use of a local representative undertaking a longhand interview. But the process would have taken longer and been more complex.
The course has brought about a more efficient use of our agents. They are able to focus on considered aspects of a claim, which also results in a less costly enquiry.
As for the insureds, no one has baulked at the interview being recorded, as it is almost expected.
Loss adjusters are perceived as old-fashioned companies, employing former police officers who undertake protracted enquiries. Their results are then presented to insurers, who then consider the liability and indemnity issues. For us, those days are over.
