There is a number that almost every car rental operator has in mind. A utilization target. In most organizations, it sits somewhere between 75% and 85%. Industry benchmarks confirm this range. Articles, consultants, and software vendors repeat it. It feels like solid ground.
But there is a problem with this number. It assumes that being 5 points above the target and 5 points below are equally bad. They are not.
The curling principle
Benoit Rottembourg, a mathematician who led the pricing transformation at Maersk, the world's largest container shipping company, uses a curling metaphor to explain why most pricing targets are set wrong.
In curling, you throw a 19-kilogram stone toward a target on the ice. You know you will not hit it perfectly. But the consequences of missing depend entirely on which way you miss. Throw too soft, and your teammates can brush the ice in front of the stone to guide it forward. You lose precision, but you have a recovery mechanism. Throw too hard, and the stone slides past the target and out of play. No amount of brushing will bring it back.
The error is not symmetrical. And neither is the response.
Pricing decisions in car rental work the same way. But most operators set their targets as if both sides of the error carried the same weight.
What 10 idle cars actually costs
When fleet utilization falls below target, every idle vehicle represents a concrete, daily cost. The lease payment does not pause. Insurance does not pause. Depreciation does not pause. A car sitting on the lot at 70% utilization when the target was 80% generates zero revenue while accumulating the same fixed costs as every car on the road.
This is painful. But it is visible, measurable, and in most cases recoverable. The operator can react: adjust rates, push a promotion, activate a broker channel, extend a rental, reallocate vehicles to a higher-demand station. The stone was thrown too soft. The team can still brush.
What 10 turned-away customers actually costs
When utilization pushes past capacity and the operator has no vehicles left, the damage is different. The customer who cannot be served does not come back tomorrow. They go to a competitor. If it happens through a broker or an OTA, the rating drops and future visibility decreases. If it happens at the counter, the operational stress cascades: no upgrades available, no flexibility, no room to absorb walk-ins or extensions.
Worse, the revenue that was lost is invisible. It does not appear in any report. There is no line item for "customers we could not serve." The operator sees a strong utilization number and assumes the week was good. It was not. The stone went past the target and off the ice.
Why the middle is not safe
The standard recommendation to target 75-85% utilization is not wrong as a range. But treating the midpoint of that range as the goal creates a false sense of balance.
Consider two scenarios for a 200-vehicle fleet over a peak week:
- Scenario A runs at 72% utilization. That is 56 idle vehicle-days across the week. At a lease cost of 20 euros per day, the direct cost of those idle vehicles is roughly 1,120 euros. Painful, but contained. And the operator had room to capture every walk-in, every extension, every last-minute booking.
- Scenario B runs at 98% utilization. The numbers look exceptional. But the operator turned away an estimated 15 to 20 customers over the week because no vehicles were available in the right category, at the right station, at the right time. At an average rental value of 150 euros, that is 2,250 to 3,000 euros of invisible lost revenue, plus the downstream damage to broker rankings and customer lifetime value.
Scenario B looks better in every dashboard. Scenario A was more profitable. The asymmetry is hidden because one cost shows up in reports and the other does not.
Non-linear costs change the equation
Rottembourg makes another observation that applies directly to fleet operations. In car logistics, moving one vehicle from station A to station B requires one driver. Moving seven or eight vehicles requires one truck, at roughly five times less cost per unit. The cost structure is not linear. Small fleet adjustments are disproportionately expensive per vehicle, while larger rebalancing moves are efficient.
This means that the cost of being slightly below target (and needing to reposition a handful of cars) is proportionally higher than the cost of being significantly below target (where a full truck rebalancing becomes viable). Revenue management is not a linear story. A 2% drop in utilization does not cost 2% more. It depends entirely on which side of the threshold you land on and what recovery options that position opens.
How to price for asymmetry
The shift is not about abandoning utilization targets. It is about asking a different question before setting them.
Instead of asking "what utilization should I aim for," the operator should ask: "which error can I recover from faster and cheaper?"
In most car rental operations, the answer is clear. Being slightly undersold is recoverable. The vehicles are there. The counter is staffed. Walk-ins, extensions, upgrades, and last-minute direct bookings can fill the gap. Being oversold is not recoverable in the same way. The lost customer is gone. The broker penalty is applied. The reputation cost accrues silently.
This means the pricing strategy should deliberately leave a small buffer of available capacity rather than optimizing every vehicle onto the road at all costs. It means resisting the urge to drop rates aggressively just to push utilization from 82% to 90% when the marginal revenue of those last points may be lower than the optionality they destroy.
Tools like Performance Hub make this visible by tracking booking pace by departure date against the same period last year. When the operator can see that a specific week is already tracking at 85% utilization with 10 days still to go, the decision becomes clear: hold rates, protect margin, keep capacity available for high-value late bookings.
And when Pricing Insights detects that a rate drop on a specific car group is accelerating bookings faster than the utilization curve justifies, it signals that the operator may be filling the fleet too early, too cheaply, removing the very buffer that would have captured higher-value demand closer to the departure date.
The real target
The best-performing operators do not aim for a single utilization number. They aim for a position on the curve where recovery options are still open, where margin per unit is protected, and where the invisible cost of turning customers away stays close to zero.
Targeting the middle feels prudent. But in an asymmetrical game, the middle is not neutral. It is just the point where you are equally likely to make either mistake, without any consideration for which mistake costs more.
The stone that is thrown too soft can still be guided to the target. The stone that is thrown too hard is gone.
We help car rental operators see both sides of the equation, the visible cost of idle cars and the invisible cost of lost demand, by station, by channel, by departure date. Book a meeting and let's look at your numbers together.
.jpg)
