Every attempt at committing fraud starts at the front office. Wouldn’t it be extremely beneficial if it were always detected there? Wouldn’t it be great if every claim were genuine? Irrefutably, that’s utopian, but there is without doubt scope for improvement here. Employing technical solutions such as ‘protective intelligence’ helps to detect possible fraudulent claims at an early stage. But let us not underestimate the human factor. Take, for example, the Dutch campaign against phishing (‘Put down the phone, don’t click, call your bank’). It has raised consumer awareness to such an extent that the financial losses have dropped significantly.
A peak in claims
You might recognize the following situation: a new, popular device is launched, such as recently the iPhone 7. Insurance companies that insure such ‘personal devices’ suddenly see a huge increase in old devices that have ‘fallen into water, come under a car, blown out of my hands’. It is likely that a large number of these claims are suspicious. It would make a huge difference if front office staff were to detect these claims.
Reason for an experiment
One of our relations came to us with the above example, wanting to know what could be done about it. First we addressed a number of questions together. Why do we want to take this on? Why does it happen? Why doesn’t our staff recognize suspicious claims? What could be done to make sure we spot these specific claims? How do we equip staff members? How do we deal with the client? And above all: what do we want to achieve?
We concluded that the front office staff had insufficient knowledge about fraud and, in particular, too little awareness of their own role in its prevention. There was a lack of knowledge about asking the right questions and verbal indicators of lies. So the organization specifically wanted to increase fraud awareness among their own staff and undergo at least one interesting learning process. In Maslow’s Four Stages of Learning these are the steps from unconscious incompetence, to conscious incompetence, and then to conscious competence.
The aim of the enterprise was to recognize suspicious claims and put them through a different process. Ultimately, this was to result in an interesting business case or at least an interesting experiment.
More awareness and more knowledge
Step 1: was a seminar, aimed at transferring knowledge about fraud recognition, fraud signs, reasons for fraud and the psychology of the offender. The matter needed to come to life.
Step 2: was a training session in interviewing techniques. Don’t ask closed questions, but use the golden Ws: who, what, where, when, why, with what and in which way? It is more difficult, but it does uncover way more information.
Step 3: was all about verbal lie detection: what is being said and how? Lie detection is highly controversial in the Netherlands. There is little trust in the existing techniques and signs and its success is often attributed to coincidence. However, I have worked with it for a long time and have often benefited from it – this was only in the private sector, as the government does not allow its use.
What we all agree on is that two thirds of our communication is non-verbal. In neighboring countries research into and training in this are taken seriously. For example, in the United States 23 non-verbal deceit factors are employed. The infamous statement of Bill Clinton about Monica Lewinsky scored on as many as 21 of these factors. Of course, scoring on deceit factors is not hard evidence, I am well aware of that. Still it is a good way to support other information.
How to recognize a liar
Lying is hard work. A liar is having a hard time, as the whole story has been fabricated, making it difficult to remain consistent. It requires concentration, which causes stress. This stress forms the basis for indicators of lies that can’t be suppressed. It influences the openness and ease with which someone talks: the person starts stammering; he clears his throat; his tongue appears to be twisted and he uses many stop words or pauses. In order to buy time, the person repeats the question (or asks for it to be repeated).
Often liars are also extremely formal in their answers and usually they give a chronological account. If your questions jump backwards and forwards through their account, they soon get stuck.
Another difference between fraudulent and genuine claims is the real experienced emotions. In the case of a lost phone, bona fide claimants more frequently emphasize the loss of photos and other personal information. Liars tend to interlard their account with terms such as ‘it’s really true, trust me, honest, cross my hart’.
Omission is also an indicator
What you often hear liars do is that, at the crucial moment, they change personal pronoun. While they start in the first person singular, they switch over when things get too close. It is more difficult to say ‘I did not do it’ than denying it in general.
In the account of what happened important parts may, for example, be skipped. ‘I went outside with my phone in my hand; it fell under the car that just drove off’. The bit about why and how the phone fell is missing.
Another deceit factor: the use of qualifiers like ‘as far as I can remember, in fact, now I think about it, in hindsight I reconstruct, I think that is how it went’. This way, they create the opportunity to contradict things later on. Stress also raises the pitch of your voice – the closer to the lie, the higher the voice.
Listen first, then write
After sharing all this kind of information, we started the training session. We did that in the workplace itself through coaching and giving directions. There were also workshops based on a selection of calls that were recorded. And it was a success. People responded enthusiastically and shared thoughts on the calls and the way the questions were asked.
In practice, we found that the employees at the call center often literally and figuratively filled in what they expected the client would say. On top of that, they regularly accepted that they did not receive an answer to one of the questions. All in all, they were more focused on the process of handling the claim than on the claim’s content. We made them aware of this and we let them ask further questions and take notice of what was being said and what not.
Stringent claims form
The last step was developing a so-called stringent claims form. On this form, we added extra phrases such as: ‘we would like to emphasize that…’, ‘in accordance with the Criminal Code….’ and ‘we will not hesitate to take legal action if…’. Sometimes in bold lettering and in a text box.
We sent this form to our selection of suspicious claims. A selection that was obviously purely subjective, based on the deceit factors. Now, it is a remarkable phenomenon, that when we Dutch people are making a genuine claim and know for sure that we are entitled to compensation, we will at the most grumble a little about such an exaggerated form. But we will still send it back. Normally, almost 100% of the forms are filled out and returned. In our experiment it was almost 30% less.
What was the reason for this? Of course, we will never find out. But this experiment has given us great insight into the possibilities of fraud detection by front office staff.