Illustration: a conceptual comparison of rental capacity, not a chart of fleet counts or results.
More revenue after a price change does not, by itself, prove that the price change worked. To evaluate car rental dynamic pricing results, compare equivalent rental periods, measure income against fleet capacity, reconcile cancellations, and explain what would probably have happened without the change.
This is Orcar's proposed measurement framework for rental operators, not a report of verified Orcar or partner revenue uplift. Our earlier five-signal pricing guide explains how recommendations can be formed. Here the question is different: what evidence would justify calling the outcome a success?
1. Separate the booking clock from the rental clock
A reservation received today may occupy a vehicle next month. Comparing orders received before and after launch measures booking intake, not the revenue generated by vehicles operating in those same weeks.
Maintain two views:
- Booking-date cohort: reservations created during a defined period, grouped by pickup date, vehicle class, lead time and rental duration. Use this to examine demand response.
- Service-date view: rental time actually delivered within the evaluation window and the corresponding revenue allocation. Use this to evaluate fleet productivity.
Record the timezone, extraction date and cancellation status as of that date. Give cohorts comparable time to mature. A recent cohort with unresolved pickups and refunds should not be presented as final beside an older, completed cohort. When a rental crosses the window boundary, allocate its time and revenue consistently rather than assigning everything to the booking day.
2. Build a scorecard with a stable denominator
Define the measures before changing prices. The following are proposed internal management definitions, not a universal accounting standard or a claim about Orcar's current dashboard.
| Measure | Suggested definition | What to check |
|---|---|---|
| Net rental revenue | Rental charges less discounts and refunds, excluding taxes and pass-through charges | Keep cancellation fees and ancillary sales separate |
| Realized revenue per rental day | Net rental revenue / delivered rental-day equivalents | Keep partial-day treatment consistent |
| Sellable vehicle-day equivalents | Sum of capacity offered during the window, using a fixed rule for unavailable time | Report maintenance, blocked time and fleet changes separately |
| Capacity utilization | Delivered rental-day equivalents / sellable vehicle-day equivalents | Numerator and denominator must use the same time basis |
| Revenue per sellable vehicle-day | Net rental revenue / sellable vehicle-day equivalents | Read beside utilization and total revenue |
| Contribution | Net rental revenue less the explicitly included variable costs | List commissions, payment fees and other included costs |
For this denominator, include both rented and still-bookable capacity. Do not subtract confirmed reservations a second time. Exclude unserviceable time under the same rule in both periods, and retain the total-fleet view so changes in availability remain visible.
Do not hide fleet shrinkage by publishing only utilization. Also show total fleet capacity and unavailable time. If sellable capacity is zero, report the ratio as not applicable, not zero. Contribution is not net profit; vehicle financing, depreciation or fixed staff costs may remain outside the chosen calculation.
3. A real example: utilization and revenue can move apart
Hertz reported first-quarter 2025 consolidated revenue down 13% year over year, average vehicles down 8%, and utilization up 240 basis points, or 2.4 percentage points. Its commentary also identified holiday timing, leap-year effects and fleet mix as factors affecting comparisons. [21]
These are Hertz's reported results, not Orcar data and not evidence of a pricing experiment. Their value here is diagnostic: higher utilization and lower total revenue can coexist when capacity and mix change. Equally, a higher average daily rate could reflect more premium rentals rather than a better price for the same car.
Keep location, vehicle class, weekday mix, lead time and duration visible. Show segment results alongside the total so a changing mix does not masquerade as an improvement within each segment.
4. Design the comparison before the rollout
HM Treasury's Magenta Book distinguishes implementation monitoring, impact evaluation and value-for-money evaluation. It recommends planning early, including collecting baseline data and establishing an appropriate comparison when estimating a counterfactual. [13]
For a rental-pricing trial, record the rule version, intended rate, applied rate and manual overrides. Then select comparable units and a primary outcome before viewing the results. A matched comparison or a phased rollout can strengthen the design, but neither automatically removes bias.
Rental fleets add a specific complication: groups can share inventory through upgrades, and a multi-day rental can occupy dates on both sides of a test boundary. A cheaper offer may move a booking from one group to another rather than create new demand. Check those spillovers before treating customer sessions or vehicle classes as independent test units.
Record other changes, including advertising, coupons, fleet additions, insurance packaging and operating hours. Report uncertainty and sample size with an impact estimate. Without a defensible comparison, describe the result as an observed change, not a causal uplift. There is no universal two-week window that makes a trial conclusive.
5. Use a publication-ready results checklist
Before publishing an uplift claim, make sure another analyst can reproduce it:
- State the decision being tested, rollout dates and evaluation unit.
- Specify booking dates versus service dates, timezone, currency and exchange-rate rule.
- Disclose cohort size, exclusions, pending bookings, cancellations and refund treatment.
- Reconcile the numerator to source records and the denominator to fleet availability.
- Show total revenue, capacity, utilization and contribution together, plus service exceptions or complaints.
- Identify the comparison, concurrent changes, uncertainty and remaining limitations.
- Obtain permission before naming a partner or disclosing commercial results.
A useful result may be “inconclusive” or “more rental days but lower contribution.” That is more actionable than an unsupported success percentage. If you operate a fleet, use this checklist to frame a discussion with Orcar about pricing evaluation. This is an invitation to discuss requirements, not a promise of uplift.
Frequently asked questions
Does fewer bookings mean dynamic pricing failed?
Not necessarily. Booking count ignores rental length, capacity and margin. Check delivered rental days, revenue per sellable vehicle-day and contribution, alongside service quality.
Can I compare this month with last month?
Yes, as a descriptive comparison. Without accounting for seasonality, fleet changes, demand mix and other interventions, it does not isolate the effect of pricing.
Is this an Orcar customer success case?
No. It is a measurement guide using public evidence and proposed definitions. It does not publish a verified Orcar or partner performance result.
Sources