How Zego's Research Reframes Risk in Level 4 Fleet Coverage

How Zego Found Level 4 Fleets Face 2.8x Higher Claim Frequency and 45% Larger Average Loss
Zego's recent analysis of commercial motor portfolios reveals stark differences between what underwriters have traditionally assumed and what actually happens on the road. The data suggests fleets classified as Level 4 - meaning high-intensity, high-mileage operations with mixed-use drivers - register about 2.8 times the claim frequency of low-risk fleets. Evidence indicates the average paid claim for these fleets is roughly 45% larger, and aggregate losses for a single policy year can swing wildly because of loss clustering and correlated exposures.

To put that in context, Zego’s dataset of thousands of policies showed that 10% of Level 4 fleets accounted for nearly 60% of the total loss cost within the Level 4 cohort https://www.theukrules.co.uk/vehicle-safety-restrictions/ https://www.theukrules.co.uk/vehicle-safety-restrictions/ in a single year. Analysis reveals that standard per-vehicle pricing models missed this concentration by smoothing risk across the whole fleet. The implication is clear: many multiple-vehicle policies hide tail risk that underwriters and fleet managers underestimate.
Four Drivers Behind Elevated Risk in Level 4 Fleet Coverage 1. Concentration of high-exposure assets
Level 4 fleets often include a cluster of high-value vehicles or a set of vehicles performing the riskiest tasks. The data suggests when multiple high-exposure assets sit on one policy, a single event - a storm, a depot incident, or a single negligent driver - can trigger simultaneous claims. Contrast that with dispersed fleets, where exposures are spread across different operational contexts and geographies.
2. Driver mix and rotation
Analysis reveals that frequent driver rotation, use of temporary or multi-hatted drivers, and high driver turnover increase both frequency and severity. Many Level 4 operations use casual drivers to meet peaks. Evidence indicates these drivers are overrepresented in loss data because familiarity with routes and vehicle handling is lower than among dedicated drivers.
3. Operational correlations and systemic events
Aggregate risk assessment must account for correlated exposures. For example, a single maintenance supplier error can affect many vehicles in a depot. Similarly, one series of poor weather days can generate clustered claims from multiple nearby vehicles. The difference between independent and correlated losses is the difference between a manageable year and a catastrophe year for a portfolio.
4. Policy design blind spots
Multiple-vehicle policies that price on per-vehicle averages or simple underwriting bands miss intra-policy concentration. Evidence indicates common blind spots include inadequate limits aggregation, weak exclusions for high-risk uses, and absence of per-driver telematics adjustments. These design choices mask tail aggregation and misalign premium with true exposure.
Why Underwriting Multiple-Vehicle Policies Often Fails to Predict Peak Losses
The data suggests one major failure mode is the use of average-based pricing. Underwriters often assume the law of large numbers will smooth out volatility across many vehicles. In practice, Level 4 fleets violate that assumption because exposures are neither independent nor identically distributed. Analysis reveals three specific reasons average-based approaches break down.
Correlation breaks the law of large numbers
Mathematically, diversification works when risks are independent. If 50 vehicles face independent probabilities of minor claims, losses will tend to average out. If those same vehicles share a common risk factor - same depot, same routes, the same driver pool - losses cluster. Zego’s research quantifies this: when correlation rises above a threshold, variance increases non-linearly, making extreme loss years more likely.
Misaligned incentives and claims inflation
Evidence indicates that when a fleet's structure makes it administratively efficient for a single claimant to bring multiple related claims, insurers see severity creep. For instance, a multi-vehicle collision at a depot can spawn property, bodily injury, and business interruption claims. Policies without clear aggregation clauses or without rigorous causation definitions invite multiplicative claims rather than additive ones.
Underuse of telematics and dynamic pricing
Contrast two approaches: static underwriting that assigns a flat rate to a whole fleet, and dynamic pricing that adjusts by driver behaviour and vehicle usage. Zego’s dataset shows fleets with telematics-informed pricing had lower claim frequency and lower aggregate loss volatility. The difference is not minor: telematics-enabled portfolios reported a 20-30% reduction in both frequency and variance in comparable cohorts.
What Fleet Managers and Underwriters Must Accept About Level 4 Aggregate Risk
What the data suggests is uncomfortable but actionable: aggregate exposure is not just a sum of insured values. It is an architecture - the way assets, people, operations, and policy language fit together. If one part is fragile, the whole structure becomes vulnerable. Below are synthesised truths that both insurers and fleet operators need to internalise.
Truth 1: Policies are system maps, not just price tags
When you buy or underwrite a multiple-vehicle policy, you are buying a map of how losses might propagate. That map needs layers - per-vehicle risk, per-driver behaviour, concentration nodes (depots, contracts), and systemic triggers. Evidence indicates underwriters who treat policies as layered systems are better at predicting tail outcomes.
Truth 2: Substitution of risk is common
Analysis reveals risk moves. If you add higher deductible options to reduce premium, you may increase the incentive to settle small claims internally, which sounds sensible, but it also changes reporting behaviour. Conversely, relaxed maintenance checks to save short-term cost can shift risk into a future period where several vehicles fail together. These substitution effects distort expected loss timing and size.
Truth 3: Data without context misleads
Zego’s work shows raw telematics streams or claim histories mean little without context. Speeding events are not equally risky across urban delivery routes and motorway logistics hauls. Comparisons and contrasts between contexts are essential; two fleets with the same speeding incidence can have very different loss profiles because of route type, cargo, and urban exposure.
6 Practical Steps to Reduce Aggregate Exposure in Level 4 Fleets
Below are measured, evidence-based steps that limit concentration, improve pricing accuracy, and reduce the probability of catastrophic years. These are operational rather than theoretical - things both fleet managers and underwriters can implement without waiting for regulatory change.
Break policies into logical risk pools
Instead of a single omnibus policy for a heterogeneous fleet, create smaller pools aligned to operational similarity: depot A urban vans, depot B long-haul trucks, temp driver pool separately. The data indicates smaller, more homogeneous pools reveal correlation patterns and allow targeted mitigation.
Introduce per-driver telematics and reputation scoring
Install or mandate telematics for high-exposure vehicles and use driver reputation in pricing and deployment decisions. Zego’s evidence indicates dynamic adjustments cut frequency and variance. Make scores actionable: restrict high-risk drivers from high-concentration tasks until retrained.
Use aggregate limits and event-based sub-limits
Policy language matters. Establish clear aggregate limits for single events and use sub-limits for high-risk exposures like loading activities or depot incidents. This reduces ambiguity when multiple correlated claims arise from one cause.
Stress-test portfolios for systemic scenarios
Run scenario modelling: depot flood, supplier failure, mass driver illness. The analysis reveals that portfolios with even small concentrations fail stress tests more often than expected. Stress-test results should feed pricing and reinsurance strategy.
Mandate proactive maintenance and supplier controls
Insist on evidence of preventive maintenance regimes and audit critical suppliers. Many Level 4 loss clusters trace back to neglected maintenance or a single faulty supplier. Treat supplier controls as part of underwriting due diligence.
Implement retroactive premium adjustments and booking reserves
Where uncertainty is high, use premium adjustment clauses or retrospective rating bands to share risk until true frequency-severity experience emerges. This aligns incentives and reduces shock to either party when atypical years occur.
Practical examples Example 1 - Depot consolidation risk
A regional delivery firm had 120 vans across two depots. After consolidating to one larger depot to save rent, a single arson attack resulted in 18 vehicle losses and a spike in injury claims from a night shift. The aggregator policy lacked a robust event aggregate limit, amplifying insurer losses. Post-event, the insurer required geographic dispersion and increased per-event limits for similar clients.
Example 2 - Driver rotation mitigation
A logistics operator rotated casual drivers between high-risk night routes and light urban runs. After several collisions involving rotated drivers, the operator instituted a driver-qualification matrix and route-specific assignment rules. Claims frequency fell and the insurer adjusted rates downward on renewal, reflecting lower correlation.
How to measure success: metrics that matter
To move from insight to verified improvement, track concrete metrics rather than hope. The following table lays out practical KPIs and what improvement looks like based on Zego's benchmarked ranges.
Metric Why it matters Target improvement Claim frequency per 100 vehicles Direct measure of loss incidents Reduce by 15-30% within 12 months with telematics Variance in annual aggregate loss Indicates tail risk and clustering Cut variance by 25% through pool segregation and limits Average paid claim (severity) Managed by maintenance, speed controls, and policies Lower by 10-20% with preventive regimes Time to notification Faster reserves and response reduce escalation Target < 48 hours for major incidents Final thoughts: treat multiple-vehicle policies as engineered systems
Zego’s research is a clear wake-up. The data suggests Level 4 fleets contain structural risks that average-based pricing will not capture. Analysis reveals successful strategies are less about clever rate tables and more about understanding how exposures interlock and how losses propagate.

Think of a Level 4 fleet like a complex building. You can measure each brick, but if they all rest on a single shaky foundation, the structure can fail suddenly and spectacularly. The right response is not just better bricks; it is foundation repair, fire-stopping, and evacuation plans. For insurers and fleet managers, that means pool design, telematics, targeted policy terms, and disciplined operational controls.

Evidence indicates those who adopt a systems approach reduce both expected loss and the probability of catastrophic years. Comparison with portfolios that cling to old, averaged assumptions shows clear financial advantage for early adopters. The path forward is pragmatic: measure, segment, mitigate, and price with humility for correlation. That is how the disturbing patterns Zego uncovered become manageable, and how multiple-vehicle policies stop being accidental time bombs.