Is your fleet’s risk concentrated in a small number of drivers?
Most organizations managing a fleet assume their risk is broadly shared across their drivers. But data tells a different story.
After years of working with mobility organizations around the world, one finding consistently stands out: a small proportion of drivers account for a disproportionately large share of insurance claims. Understanding where risk is concentrated – and which drivers carry the most of it – is one of the most valuable things a fleet manager can know.
Here, Johan Forseke explains how risk is typically distributed across a fleet, how AI makes it possible to identify it, and what to do with that information once you have it.
How is driver risk distributed across a fleet?
Driving risk usually develops over time, shaped by habits, patterns and behaviors that build gradually. Because of this, it means risk can be detected as it builds – even before it would trigger traditional event-based alerts.
A driver who never exceeds the speed limit and never triggers a harsh brake event can still be identified as high risk. That’s because the behaviors that are indicative of crashes aren’t always the obvious ones. They might be subtle deviations from “the norm” that are only visible when you’re able to measure the right thing. Read how behavioral patterns predict crash outcomes.
What does Greater Than’s data show about risk concentration in fleets?
Independent validation of Greater Than’s AI has reinforced just how concentrated fleet risk can be.
The validation confirmed a 25x differential in crash probability between the safest and riskiest driver profiles. This means that drivers categorized as highest risk are 25 times more likely to crash than those identified as lowest risk. This is a significant difference.
But not just that. The validation also confirmed that drivers identified as highest risk by our AI are 91 times more likely to be involved in a crash associated with a severe or major claim than those identified as lowest risk.
This has huge implications for fleet risk management. It shows that risk mitigation efforts don’t need to be applied in the same way across the board. In fact, it is much more effective to concentrate efforts where they have the potential to prevent the most serious, and costly, crashes.
How can AI identify which drivers are high risk?
While traditional telematics scores measure events that require a driver to pass a certain threshold, AI is different.
Greater Than’s pattern-based AI continuously analyzes driving behavior throughout entire trips, identifying patterns that it compares against a database of billions of real-world driving profiles linked to known outcomes. This makes it possible to calculate which drivers are most – and least – likely to be involved in a crash, based on the behavioral patterns the AI has seen before.
The result is a risk ranking across an entire driving population; rather than a list of drivers who have triggered event alerts. Watch a video to see how pattern AI works.
What can fleet managers do once they know where risk is concentrated?
The identification of the highest-risk drivers is just one part of the overall picture. It’s what fleet managers do with the information that matters most.
Typically, I see this type of actionable intelligence used to prioritize coaching or training for the drivers who need it most. This means that, rather than roll out training to all drivers, organizations can save time and resources by targeting the top 5% or 10% highest-risk drivers, for example. This is likely to have a greater impact on claims than applying training more broadly. Read more about using AI to prioritize driver training.
Does risk concentration look the same across different types of fleet?
All organizations are different, so the exact distribution of risk will vary.
However, as our AI validation has confirmed, drivers identified as higher risk are consistently associated with a disproportionate share of claims – across multiple continents, different vehicle types and different driver categories. This means that, while the risk concentration might vary, the principle behind it doesn’t.
