Watch: Applying Data Science to Trucking Insurance
Data science is bringing innovation to trucking insurance underwriting, says Ian White, managing director of Koffie Labs.
The underwriting process for trucking insurance is relatively antiquated, having not changed in the past 50 years, White says.
Understanding drivers is one of the main considerations licensees, or underwriters, have. “You look at their motor vehicle record for any type of disqualifying events,” White says. “It could be an accident or violations. And they take into account the operations of the road haulier itself, which includes things like financial stability, the type of goods they are carrying, where they are based and whether they are long-distance haulers. . That’s what the incumbents have been doing for about 50 years or more.
Truck insurance is no small business. White estimates that “moderate” underwriting represents about $100 billion in premiums. “We could define it more broadly and get closer to maybe $150 billion, maybe $200 billion. For reference, personal auto insurance accounts for $250 billion in premiums per year.
Continuing his criticism that the industry lags behind in innovation, White says licensees don’t consider equipment when signing up for a fleet. A fleet of 1999 Western Star trucks will get the same quote as a fleet of top-of-the-line 2022 Volvos.
The data must be analyzed to truly understand how these fleets perform differently. This means not only looking at accidents and violations, but also the vehicle equipment, engine manufacturer and whether operators use technologies such as telematics. “It gives us a much more granular and dynamic view of risk,” says White, “not just today but in the future when technology will play an increasingly important role in driving a vehicle as that the role of the driver will begin to diminish.”