Usage-based insurance in personal motor is a story that has largely been told. The telematics-enabled product found a stable market position: younger drivers with limited claims history pay for continuous risk observation as a mechanism to access pricing that reflects their actual driving behavior rather than their demographic profile. The actuarial case was clear early. The technology has matured. The product works.
Commercial motor fleet insurance is where the interesting math is now. The market is structurally different, the data richness is significantly higher, and the gap between traditional flat-rate pricing and actuarially sound usage-based pricing is wider than in personal lines — which means the correction, when it comes, will be larger.
Why Commercial Fleet Was Left Behind
Personal lines telematics made sense first for two reasons: the customer base is large enough to fund the technology investment, and the underwriting decision is essentially binary at the policy level (one vehicle, one named driver, one annual period). The policyholder decides whether to participate; the carrier decides whether to accept; pricing adjusts based on observed behavior.
Commercial fleet pricing is more complex in ways that initially made telematics harder to implement. A fleet of 40 vehicles involves 40 drivers of varying skill and tenure, driving schedules that shift with business demand, cargo that varies by shipment, and multiple vehicle types with different risk profiles. The insurer must underwrite the fleet as a portfolio, not as 40 independent policies. The actuarial model is a fleet-level loss rate, not an individual driver loss rate.
Traditional fleet underwriting addressed this complexity with a rough proxy: fleet size, industry sector, claims history, and vehicle age distribution. A courier fleet with 10–50 vehicles and three years of clean claims history gets rated on those five inputs. The model uses the observable to proxy the unobservable — driver behavior, route risk, cargo handling practices — because the insurer had no access to the actual unobservable variables.
Telematics changes this. When every vehicle in the fleet generates continuous telemetry — GPS track, speed profile, harsh braking events, idling time, geofence violations — the unobservable becomes observable. The actuarial model can work directly on the risk drivers rather than on the proxies.
The Actuarial Math in Commercial Fleets
The premium differentiation potential in commercial fleet telematics is larger than in personal motor. In personal motor, the risk ratio between a high-risk and low-risk driver is typically in the range of 3:1 to 5:1 on expected loss cost. In commercial fleet, the equivalent differentiation is wider — partly because fleet size creates a longer-tail risk distribution, and partly because commercial vehicles have higher loss severity per incident.
A concrete example: a logistics operator running a 30-vehicle refrigerated goods fleet. Traditional pricing uses industry sector, claims history, and vehicle age. But within this fleet, telemetry reveals that night-shift drivers on the northern regional route have a harsh-braking event rate 2.8× higher than the fleet average, and that three specific vehicles with high odometer readings are showing maintenance anomalies correlated with increased incident probability. The flat fleet-rate pricing is cross-subsidizing the high-risk subset. A usage-based pricing model — either premium adjustment at renewal based on observed behavior, or real-time micro-adjustment for fleet managers — would price these risks correctly.
For the insurer, the pricing correction improves combined ratio on the fleet book. For the fleet operator, the behavioral data creates an operational incentive structure that reduces incidents beyond what the insurance adjustment motivates. The telematics product is not just an insurance pricing tool; it is a fleet risk management tool, and the operator's interest in reducing incidents aligns with the insurer's interest in reducing claims.
The Gig Economy and Dynamic Fleets
The most acute version of the usage-based commercial insurance problem is in gig-economy fleet structures: last-mile delivery operators, ride-hail vehicle owners, and similar models where vehicles are active for variable periods under commercial use and inactive under personal use. The risk profile of a vehicle shifts dramatically between these two usage modes, and traditional annual policy structures price the commercial risk across the entire 8,760-hour year regardless of actual commercial use patterns.
This is where Zego's approach — which we backed in 2021 — addresses a genuine actuarial gap. The per-mile or per-hour commercial cover product prices the period of commercial use separately from personal use, which is actuarially correct and commercially sensible for operators whose vehicles are not in commercial service full-time. The pricing is technically more complex because it requires managing period-level risk rather than annual-aggregate risk, but the data to support that pricing is available through the telematics stream.
The regulatory environment for this type of product is still evolving in several EU jurisdictions. In some markets, the distinction between personal and commercial motor use periods is not yet cleanly handled in policy wording. That creates a short-term friction but a medium-term moat for the operators who have worked through the regulatory alignment ahead of market normalization.
What We Are Watching
The next wave in commercial fleet telematics is not better sensor hardware. It is integration with fleet management software, dispatch systems, and cargo tracking platforms that are already generating the operational data insurers need. The insurer does not need to install new hardware in every vehicle; the fleet operator's existing telematics infrastructure already contains the relevant risk signals. The question is how to access that data in a structured, consented, and actuarially usable format.
We are not saying the telematics hardware market is dead — for smaller fleets without existing digital infrastructure, hardware installation remains necessary. But for the growing segment of commercial fleet operators who already run sophisticated telematics for operational reasons, the insurance pricing opportunity is in data integration, not data collection. That is a different technical challenge and a different sales motion, and it is closer to the SaaS world than to the insurance product world.
The companies building those integration layers — connecting existing fleet telematics to insurance pricing APIs — are building something that compounds as fleet digitization increases. That is the direction we are most interested in for the next cycle of commercial motor investment.