How to Compare Car Rental Deals: Providers, Vehicle Types, and Prices Across Locations

A line of different rental vehicles from compact to SUV on a sunny lot with airport terminals in the background

The cheapest rental car at Heathrow right now is not listed on the first site you check. It is on the third, or the fifth, or a provider you have never heard of. That is not a guess — that is how aggregator search works. Different agents negotiate different net rates with the same fleet. The vehicle is the same Toyota Corolla at the same Hertz desk. The price is not.

This page covers what a car rental comparison actually returns: providers, vehicle types, pickup locations, and pricing patterns — and why searching one site is the most expensive way to rent a car. The car rental API post covers the technical contract. The car rental aggregator API post covers how Ktemia fans out one search to many providers. This one covers what the search surfaces and why it matters.

Key Takeaways

  • The same car at the same desk costs different amounts through different providers. Aggregators surface that gap.
  • Vehicle categories follow ACRISS/SIPP codes: Mini through Luxury, SUV, Van. The make and model depend on the depot’s fleet that day.
  • Airport pickups carry concession surcharges (often 8–12% in the US). Downtown and dealer depot locations skip them.
  • Mileage policy (unlimited vs. capped), fuel policy (full-to-full vs. prepaid), and insurance inclusion vary by provider for the same car class.
  • Ktemia compares across aggregators. One search, many agents, same pickup. Not live yet.

Why the same rental car costs different prices

A rental car price is not set by the company that owns the car. It is set by the agent — the booking platform that brokered the deal. Hertz sets a rack rate. Booking.com negotiates a net rate. Expedia negotiates a different one. A small OTA you have never visited negotiates a third. The car is a Hyundai Tucson at Terminal 5. The price varies by who sold it.

Skyscanner’s Car Hire docs (retrieved 25 September 2026) define this split explicitly: vendors own the cars (Hertz, Sixt, Enterprise). Agents are supply partners that provide quotes. A vendor can also be an agent — Hertz sells directly and through aggregators. When you search on Skyscanner, you see quotes from multiple agents for the same vendor’s car.

This is why a comparison search returns five prices for the same vehicle class at the same pickup. They are not duplicates. They are different agent contracts for the same fleet.

The price gap between the cheapest and most expensive agent for the same car class, same pickup, same dates is routinely 20–40%. On long hires (2+ weeks) or in high-demand markets (summer in Southern Europe, ski season in the Alps), the gap can exceed 50%. People who book from the first tab they open are paying the difference without knowing it exists.

What vehicle types does a car rental search return?

Car rental search results are categorised using ACRISS codes (also called SIPP codes) — a four-character standard that every major rental company and aggregator uses. The first character is the vehicle category:

Code Category Typical vehicles
M Mini Fiat 500, Toyota Aygo
E Economy Toyota Yaris, Hyundai i10
C Compact VW Golf, Toyota Corolla
I Intermediate Hyundai Tucson, Kia Sportage
S Standard Toyota Camry, VW Passat
F Full-size Ford Mondeo, Chevrolet Malibu
P Premium BMW 3 Series, Mercedes C-Class
L Luxury BMW 5 Series, Mercedes E-Class
X Special Convertibles, sports cars
J SUV RAV4, Nissan Qashqai
V Van / Minivan Ford Galaxy, VW Transporter

The second character is the door type (2-door, 4-door, wagon, SUV). The third is transmission (M = manual, A = automatic). The fourth is fuel/AC (R = AC with petrol, D = AC with diesel, E = AC with electric).

The make and model in the result is not guaranteed. Booking.com’s docs and Hertz’s own booking flow both say “or similar.” You book a category, not a specific car. The Corolla in the search result might be a Civic at the desk. Same class, different badge. This is standard across the industry and is not a bug in the API.

Most comparison sites show a photo of a specific car (a shiny Corolla) but the booking terms say “Compact, or similar.” If you are building a comparison UI, do not promise the exact make. Show the ACRISS category and say “or similar.” Every major vendor’s terms of service requires this. Displaying a specific model as guaranteed is how you earn a chargeback.

How do providers and agents differ?

A provider is anyone who appears in the search results with a price. But not all providers are equal:

Fleet vendors own the cars. Hertz, Sixt, Enterprise, Europcar, Avis, Budget. They set the base rate, the mileage policy, the fuel policy, and the insurance terms. When you pick up the car, you are dealing with the vendor regardless of who you booked through.

Aggregator agents broker the deal. Booking.com, Rentalcars.com (owned by Booking Holdings), Skyscanner, Kayak, Expedia, AutoEurope, CarTrawler. They search multiple vendors and return quotes. Their price may be lower than the vendor’s direct site because they negotiated a net rate. It may also be higher because they added a margin.

Broker agents are smaller OTAs and travel agencies that resell aggregator inventory, sometimes through white-label platforms like CarTrawler. They show up in Skyscanner results as agents you have never heard of — but the car at the desk is still Hertz or Sixt.

Car dealer car rentals are a fourth category. Dealerships that partner with Enterprise or regional fleet companies offer vehicles through the aggregator supply chain, often listed under the partner brand with the dealership address as the depot. More on this in the car rental API post.

When you compare across providers, you are comparing the same fleet through different middlemen — each with a different margin, cancellation policy, and insurance bundle. The car is the same. The deal is not.

How does pickup location affect car rental prices?

Three location types appear in aggregator search results:

Airport (in-terminal or shuttle). The most common pickup. Also the most expensive — airports charge rental companies a concession fee for operating on-site, and that fee is passed to you. In the US, airport concession recovery fees typically add 8–12% to the base rate. In Europe, VAT is already included but the airport premium is still baked into the rack rate.

Booking.com’s Demand API uses depot_location_type to distinguish in-terminal pickups (walk from baggage claim) from shuttle locations (bus ride to an off-airport lot). Shuttle depots are usually cheaper — same vendor, lower concession fees.

Downtown / city depots. No airport concession surcharge. Shorter operating hours. No shuttle. These are the depots that show up when you filter by “downtown” or “other locations.” If you are staying in the city and do not need the car until day two, a downtown pickup can save the 8–12% airport tax on every day of the rental.

Dealer and suburban depots. Enterprise at a Ford dealership, Hertz at a body shop, a local fleet company in an industrial park. These are the cheapest locations on paper — no airport tax, no prime retail rent — but they may not appear in every aggregator search because the depot’s opening hours or fleet size is too small for the aggregator’s minimum threshold.

The same Economy car on the same dates at LHR Terminal 5 versus LHR Shuttle versus Central London can show a 15–25% price spread. The car is the same class. The vendor may be the same. The depot type changes the concession fee, the operating hours, and sometimes the insurance bundle. A comparison search that only shows airport in-terminal results is hiding the cheaper option three miles away.

What else varies between providers for the same car?

Beyond the price itself, four things change between providers for the same vehicle category at the same pickup:

Mileage policy. Unlimited kilometres or a daily cap (often 200–300 km/day in Europe). Going over the cap incurs per-km charges, typically €0.20–€0.50. Booking.com’s Demand API exposes mileage_type as an optional filter. If you are driving from London to Edinburgh, a 200 km/day cap will cost you more than the headline price suggests.

Fuel policy. Full-to-full (return with the same fuel level) or prepaid (the vendor fills it, you pay upfront at a premium per-litre rate). Full-to-full is almost always cheaper if you can refuel near the depot before drop-off. Prepaid fuel is a convenience surcharge.

Insurance inclusion. Some agents bundle Collision Damage Waiver (CDW) and Theft Protection (TP) in the headline price. Others show a lower base rate but add insurance at the counter. The €15/day car with CDW included is cheaper than the €12/day car that charges €18/day for CDW at pickup. A comparison search that surfaces the insurance terms alongside the price is the only way to see this.

Cancellation policy. Free cancellation up to 24–48 hours before pickup is standard on most aggregators. But some agents require 72 hours. Others charge a cancellation fee regardless. This matters if you are booking early to lock a rate and may need to change plans.

A true comparison is not “cheapest headline price.” It is cheapest total cost after mileage, fuel, insurance, and cancellation terms. The car rental API contract returns these fields when the providers include them in their feed. It cannot invent them.

What Ktemia compares — and what it does not invent

Ktemia’s car rental search is POST /v1/cars with pickup, dates, and driver age. The route is not live yet. When it is, the search will return:

  • Vehicles by category — ACRISS/SIPP class, not a guaranteed make and model.
  • Prices from multiple providers — different agents, same fleet, different deals.
  • Pickup locations — airport, downtown, dealer, and suburban depots.
  • Policy fields — mileage, fuel, insurance, and cancellation where the provider includes them.

What it will not do:

  • Invent prices. If the route is not live, there is no response — not a fake one.
  • Guarantee a specific vehicle. “Compact, or similar” is the industry standard.
  • Book the car. Search is comparison. Booking is a different job. That is in the terms.

The same search runs in the dashboard, the API, and the copilot. If pickup or dates are missing, the API rejects the call and the copilot asks. Neither guesses. That contract matches flights and hotels — same platform, same discipline.

On Free, you get 50 API requests a month at $0. Pro is $29.99/month with 20,000 requests at 20 per second. Business is $299 with 250,000 at 250 per second. The monthly cap is one shared pool — car searches and flight searches draw from the same number.

Compare across providers, not within one.

The first price you see is not the best price. It is one agent’s margin on one vendor’s rate at one depot. A comparison search queries many agents, returns many prices for the same car class, and surfaces the mileage, fuel, and insurance terms that make the difference between a cheap headline and an expensive trip.

Ktemia’s car search runs across aggregators — same contract in the dashboard, the API, and the copilot. Routes are not live yet. When they are, this page will link to a dated response with real providers, real categories, and real prices. Not invented Hertz rates.

We published the underlying numbers too: rental car prices against customer ratings across 13 rental companies, so you can check the working rather than take our word for it.

Start from the car rental API for the technical contract, or pricing for the caps.

Sources

  1. Booking.com, About Car Rentals (Demand API) (retrieved 2026-09-25)
  2. Booking.com, Search for car rentals (retrieved 2026-09-25)
  3. Skyscanner, Car Hire Live Prices overview (retrieved 2026-09-25)
  4. Skyscanner, Car Hire Live Pricing API (retrieved 2026-09-25)
  5. Expedia Group, Rapid Car API (retrieved 2026-09-25)
  6. ACRISS, Car Classification Codes (retrieved 2026-10-02)
  7. Ktemia, Pricing (retrieved 2026-09-25)
  8. Ktemia, Terms of Service (retrieved 2026-09-25)

FAQ

Why are rental car prices different on every site?

Each site is a different agent with a different contract. Hertz may give Booking.com one net rate and Expedia another. The vehicle is the same Corolla at the same desk. The price is not. Aggregators search multiple agents. A single site searches one.

What vehicle types show up in a rental car search?

Most aggregators return ACRISS/SIPP codes that map to categories: Mini, Economy, Compact, Intermediate, Standard, Full-size, Premium, Luxury, SUV, Van, and Convertible. The exact makes and models depend on the fleet at that depot on that date.

Is it cheaper to rent from an airport or a downtown location?

It depends on the market. Airport locations charge concession fees (8–12% surcharge at many US airports). Downtown depots skip that fee but may have shorter hours and no shuttle. A comparison search shows both so you can see the actual price difference.

How many providers does a car rental comparison search?

Booking.com Demand API says it searches its own connected suppliers. Skyscanner Car Hire returns quotes from multiple agents and vendors. Expedia Rapid claims 47,000 vendors. The number you see depends on the aggregator, the pickup location, and the date. A comparison that searches across aggregators returns the widest set.

Can I filter rental cars by transmission, fuel type, or mileage policy?

On Booking.com Demand, yes — transmission_type, fuel_types, mileage_type, and air_conditioning are optional filters. On Skyscanner Car Hire, filters depend on the result set. On Ktemia, the contract will accept optional filters once the route is live. The required fields are still pickup, dates, and driver age.

Stop switching tabs to find a better deal.

One search. Flights, hotels, cars, and real estate — compared instantly.