For a company whose expressed strategy is to provide ‘sublime’ client service, as expressed by Director of Claims, Brian Wahl Olsen, one thing Alm. Brand never wants to do is to wrongfully accuse its policyholders of deceiving.
But weeding out the fraudulent claims, especially in an era of increasing digital self-service, is of outmost importance in order to protect honest customers’ sense of justice, price levels, as well as the profit margins of one of Denmark’s largest and most well-established insurance companies.
“In order to protect the vast majority of our honest clients against the fraudulent practices of the few, we have introduced SAS® Fraud Framework. We always assume that all customers are honest − but at the same time, we know that we reveal fraud to the tune of more than EUR 5.3 million annually. We are more and more dependent on self-service to let our clients file their insurance claims at the time and place which suit them best. To make sure that we can still find the fraudsters even when people are filing claims online, we are dependent on strong processes and systems,” says Brian Wahl Olsen.
A man who had reported a lot of valuable electronics and jewelry stolen from his home only to be revealed selling similar items online told investigators that he came home to find the items in a box: Probably the thief felt guilty and returned the items, he explained. Brian Wahl Olsen Director of Claims Alm. Brand
Dating back to 1792 and established by royal decree by the Danish King Christian VII, Alm. Brand was established to secure citizens and business owner’s access to fire insurance for their buildings. Today, the company has 1600 employees, a revenue of approx. EUR 1 billion and offers a large number of insurances to individuals and businesses.
Detecting fraud is ‘basic police work’
Brian Egested heads up Alm. Brand’s Investigative Unit. When his team of investigators is called upon to look deeper into the circumstances surrounding an insurance claim at Alm. Brand, it happens because someone has raised a red flag. Sometimes it is a claims officer with a keen eye for the unusual, in some instances an anonymous tip. Over the last year, it is increasingly the insurance company’s fraud detection system developed by SAS Institute, which alerts the claims investigators that something seems awry.
“When we get an alert, we quickly gather as much information as we can. We speak to witnesses and make surveys of the location; for example date and time of a house fire or car accident, and provide information on the claimant. It is just regular police work, and often something we do in direct collaboration with the police,” says Brian Egested. As a former special investigator in the Danish National Police force, he is no stranger to the job.
A man claimed as part of an injury insurance claim to be visually impaired at the level where he could only detect shadows.
A surveillance officer reported seeing the man effortlessly backing a car with a trailer attached out of a shopping center parking space. Brian Wahl Olsen Director of Claims Alm. Brand
Weeding out the fraudulent claim
Finding the fraudsters and clearing upstanding customers of suspicion quickly is of utmost importance. Brian Egested explains that when using the SAS® Fraud Detection framework, the evidence builds already as the claim is being telephoned in or added to the self-service claims systems on the company’s website. The system is quickly able to compare the history of the claimant and various aspects of the circumstances of the incident to gauge whether the case is one that should be forwarded to the investigation unit.
“If a young man crashes a luxury car at four in the morning on a deserted city street, which then bursts into flames with no witnesses and he had a similar claim two years ago, it is perhaps worth looking into. But sometimes, circumstances are not as obviously suspicious,” says Brian Egested.
He refers to a specific case in which he is convinced that the fraud was only revealed due to the SAS system for alerting. A man had three different claims with the company at once and came up as an alert and as the investigation deepened, it turned out he was insured in no less than seven companies. He was ultimately denied coverage as he had lied in his application form about previous claims.
Brian Wahl Olsen is confident that Alm. Brand can continue to build on its promise of sublime customer service while turning away those who try to take illegal advantage. Building on the experiences from Fraud Analytics, he wants to expand the focus to make customer experiences better for the honest majority.
“We want to become much more analytical in our approach by getting more information about what customers really want from their relations to Alm. Brand. We want them to feel recognized no matter how they choose to contact us,” Brian Wahl Olsen concludes.
A person tried to file a claim for large collection of gold jewelry, which was reported stolen.
When investigators ran a picture search online against the pictures sent in as proof of ownership, identical images of jewelry turned up on Google from various online sellers Brian Wahl Olsen Director of Claims Alm. Brand
Will we ever get to the point when we can spot the fraudster from the very outset?
Brian Egested says: Probably not, unless they have a track record and people deserve the benefit of the doubt. Research has been conducted which shows that perhaps 10 per cent of people have criminal tendencies and 10 per cent are always honest regardless of the circumstances. The remaining 80 per cent may get tempted to break the law and could become insurance fraudsters if confronted with economic adversity and personal problems. Nevertheless, it is getting harder to get away with, not least due to data analytics, and often, people are relieved that they were not successful and made themselves criminals by receiving a compensation to which they were not entitled.
- Article A guide to machine learning algorithms and their applicationsDo you know the difference between supervised and unsupervised learning? How about the difference between decision trees and forests? Or when to use a support vector algorithm? Get all the answers here.
- Series Meet the data scientist: Kristin CarneyWhen Kristin Carney graduated with a BS in mathematics, she wasn't sure what she wanted to do with her degree. That’s when she began researching data science.
- Article Proactive detection – A new approach to counter terrorTo counter terror, investigative teams can better utilize the data they already have by applying a fresh approach with these steps to proactive detection.
- Series Evolving your institutional insightBig data in education can help administrators understand college applicants and students – including which ones will struggle and why.