How we measure car hire

Where the data comes from

We search 25 airports on a schedule through DiscoverCars, who aggregate the suppliers. Each sweep prices the same trip at four rental lengths — a week, a fortnight, a month and 45 days — because charges that are levied per rental look completely different as a share of the total depending on how long you hire for.

Searches are run 45 days ahead: far enough out to be a normal booking, close enough that inventory is real. All prices are quoted in euro for comparability, from a UK-resident, 35-year-old driver profile. Residence and age genuinely change what suppliers offer, so holding them fixed is what makes two markets comparable.

Every result is appended to a log. Nothing is overwritten, so a market's history accumulates and its figures get more reliable over time rather than jumping about.

How each figure is calculated

Suppliers charging for a second driver

Of the offers that publish an additional-driver fee, the percentage whose fee is above zero. Offers that do not publish a fee are excluded from the denominator, not counted as free — assuming silence means free would overstate how generous a market is.

Typical second-driver fee

The median fee across all offers that publish one, including those that publish zero. It is the fee for the whole rental, not per day: one fee per rental however many drivers you add. Where we show it as a percentage, that is the median fee over the median rental price for the same search.

Typical deposit

The median amount the supplier states it will pre-authorise on the main driver's card. A pre-authorisation is a hold, not a charge, and is released after the car comes back — but it has to fit on the card. Where we show banding (high, average, low), that is the supplier's own classification, not ours.

Cancel free

The share of offers the supplier marks as cancellable without charge.

Collection type, fuel policy, mileage, cross-border, debit card

Straight shares of the offers in a search. "Free shuttle service", "In terminal", "Same to same", "Full to full" and the deposit bands are the suppliers' own labels, reproduced rather than interpreted.

Median week, median month

The median advertised rental for that duration. Rental only — this is the number we exist to argue with, not the number we recommend deciding on.

Turning several sweeps into one figure

A market report shows the median across every sweep of that market, not the latest reading. One search is an anecdote; the median across repeated searches is the market. Each report states how many searches it is built on, and a market is not published at all until it has enough of them.

Advertised price versus what it really costs

The advertised price is the rental. What you pay can also include a second-driver fee, an excess-cover product, fuel policy differences, and a charge for exceeding an included mileage cap. Our search tool adds the second-driver fee to every price the moment you say you need one, and prices fuel and mileage where you tell us the distance.

Excess cover is deliberately not folded into our totals. Plenty of travellers already hold annual excess insurance or rely on a credit card, and silently assuming a purchase would be the same mistake this product exists to expose. We show what cover would cost beside the rental instead — at some airports it is more than the car.

Mileage overage is never in a total either: the supplier sets that rate and does not publish it to us, so we flag the cap rather than guess the charge.

What we will not claim

• Nothing we have not measured. No terminal numbers, shuttle timetables, desk addresses or opening times beyond what the suppliers state in the data. A wrong terminal number on an otherwise careful page destroys the trust the measured numbers earn.
• No seasonality yet. Comparing months requires months of history. When we have it we will say so, with the dates.
• No prices as facts. Car hire is priced dynamically. Everything here is a typical value from a stated window — the only way to know your price is to search your dates.
• No rankings by commission. Results are ordered by price. We earn a commission when a booking completes and the traveller pays nothing extra; see our affiliate disclosure.

Freshness, and what we publish

Each report carries the window its data was collected in and the number of searches behind it. We do not touch those dates to imply freshness — if a market has not been measured recently, its page says so by showing you the real window.

We publish aggregates: shares, medians and counts. The underlying per-search records stay ours. You are welcome to quote the aggregates with attribution — they cannot substitute for a live search anyway, because nobody books a car on a median.

Corrections and questions: hello@carrentalbrain.com. If you think a figure here is wrong, tell us and we will check it against the log.