The insurer is taking a new, comparative approach 

AXA UK claims handlers are piloting a machine-learning tool to support the firm in making quicker and more accurate decisions regarding motor claims.

The tool, Repair or Replace Intelligence (RoRI), captures relevant data from phone conversations using machine learning. It then assesses how much it would cost to fix or write off vehicles, aiming to make the right decision as soon as the claim is notified.

The pilot starts today (10 February) in AXA’s Ipswich office, and the plan is to integrate RoRI into the Guidewire software used by AXA later this year.

Waseem Malik, executive managing director, claims, said: “Claims analytics is at the centre of our business strategy, which relies on data-driven insights. Our ambition is to have seamless interactions with our customers. We believe that by empowering our people through technology, we will get closer to our customers.”

This development follows AXA rolling out three artificial intelligence (AI) bots last year

Comparative approach

RoRI collects and uses relevant information and, based on a group of predictive models, it then compares the cost of repair with the cost of total loss.

The former is affected by the type of damage, the parts needed, the labour required and whether the work is being done by an approved repairer.

The latter, on the other hand, is usually the replacement cost minus salvage.

AXA UK is taking a comparative approach with this model, which differs widely from its previous strategy. This only considered the age of the vehicle and the extent of damage.

This meant that in that decision matrix, only half of total losses were identified immediately at First Notice of Loss (FNOL) and the other half was identified once the vehicle was already at the garage.

This, in turn, led to delays for customers - this had to be tackled to improve service and, from an insurer perspective, it also generated unnecessary storage and credit hire costs.

Smoothing it out

AXA UK also decided to smooth out the motor claims process by providing better support to claims handlers.

Its focus was on capturing more relevant data, but without making calls longer or adding any complexity.

It turned out that the phone conversations with customers were much richer than the previous matrix assumed.

Michael Calvert, head of claims analytics, added: “For a car owner, it can be frustrating to wait for your vehicle to be repaired only to find out later that it actually needs to be written off. We want our customers to have a smooth claims journey.

“The best way to do that is to innovate and to support our employees. By giving them this machine learning tool, we trust they’ll be able to reach the right decision more quickly.”