Charles Smith, VP of product at SambaSafety, explains why fleets must leverage data for better risk predictability.

If you can’t decide what to wear today, how about watching the weather forecast? That “70% chance of rain” prediction might not guarantee getting caught in a shower, but you can, at least, prepare by putting a raincoat on. We trust the Met Office to give us a pretty reliable idea of the prevailing conditions via tried-and-tested measuring equipment and established understanding of weather patterns, providing sophisticated visualisations of the next cold front.

What if fleet managers could do the same with road risk? As it stands, many tend to chalk up every minor collision to bad luck. Trouble is, there’s no such thing as a minor collision nowadays. Increased cost of parts and long lead times for repairs have driven up the severity of crashes in financial terms, making the outlook feel overcast when it’s time for insurance renewal.

Risk isn’t random

SambaSafety recently released its 2025 UK Driver Risk Report, offering a detailed snapshot of the current climate of motor risk. Based on long-term studies of data, from UK fleets and global logistics operators who use our data aggregation capabilities, the report’s findings were eye-opening and demonstrate just how predictable risk can be.

For example, our data shows that, among employees who drive for work, crash risk peaks when they are two to three years into their role, not just when they’re new and inexperienced. Employers may be offering worthwhile safety training when onboarding new drivers, only to have complacency or overfamiliarity override those efforts later on.

Another key predictor of future incidents is the seriousness of previous speeding offences. Drivers clocked travelling at more than 20mph over the limit are shown to be 80% more likely to make a claim in the following 12 months. Having access to this data and knowing how it constitutes a ‘weather warning’ can aid managers in identifying and averting future incident risk.

Insurers are listening

Telematics data, combined with the full range of risk-related driver information, provides insurers with a powerful barometer for future claims. In their ultra-competitive marketplace, being able to make robust predictions informing appropriate pricing, without undue reliance on a fleet’s claims history or industry benchmarks, gives them a significant edge. What’s more, it allows them to advise and even incentivise fleets to create training strategies that help reinforce and perpetuate positive trends.

Our research found that 74% of fleets are already using telematics to inform driver training activities, but only 30% are sharing data with their insurers, the majority saying they’ve never been asked to. Amid the spiralling pressures on motor insurance costs, data-led risk assessment is becoming ever more essential in driving down loss ratios. Some 64% of insurers say they either already offer telematics-based insurance programmes for fleets, or plan to roll these out in 2026, so fleets need to get ready.

A brighter forecast

At SambaSafety we’ve known for a while that this is the direction the industry is taking and we’re continually working with insurance partners to encourage fleet clients to share their risk data for mutual benefit. These collaborations are where we’re seeing the biggest successes. One major UK retailer, over a five-year period of huge expansion in the size of its dotcom fleet, achieved a 22% reduction in claims frequency and a 50% reduction in third-party claims involving bodily injury, resulting in a 24% reduction in spend on third-party claims.

We know it’s challenging to make sense of telematics data alongside drivers’ history of endorsements, claims, PCNs and NIPs. Our ‘Risk Cloud’ has been designed to tackle that for you. By pulling all those risk indicators into a single view, we help you and your insurers to visualise the emerging patterns, target training efforts wherever the cloud is most ominous, and to know you’re getting it right when you see the sun shining through.