Buying a used car is one of the largest purchases most people make without a survey, a solicitor or a cooling-off period. You get a test drive, a conversation with a stranger, and whatever paperwork they choose to show you.

And yet almost everything you need to judge the car is already written down, publicly, for free — and most buyers never look.

I run a vehicle history service, which means I spend my working life looking at the records behind used cars: manufacturer service data, MOT results, mileage readings. We recently analysed our own dataset — thousands of UK vehicles that somebody paid to have checked — and the results say something uncomfortable about how used cars are sold in this country.

„Full service history” is a claim, not a fact

Across 2,966 UK vehicle lookups, 37.5% returned no digital service history at all. Not a patchy record. Nothing.

That figure needs an honest caveat, and I would rather give it to you than have you find it later. A blank result does not automatically mean the car was neglected. It can mean the owner used an independent garage that never fed records into the manufacturer’s system. It can also mean the manufacturer simply does not expose its service data to the systems the trade can query — and that varies enormously by brand. In our data, 96% of BMWs returned a digital record. Ford managed 11%.

So the practical lesson is not „walk away from any car without a digital history.” It is subtler and more useful:

A missing service history means very different things depending on the badge. On a BMW or an Audi, where nineteen in twenty comparable cars do have retrievable records, a blank is a genuine warning sign worth asking hard questions about. On a Ford, a Hyundai or a Toyota, a blank is normal and tells you very little — so ask for paper records or garage invoices instead, and don’t let a seller talk you into paying a premium for a „documented” history that was never digital in the first place.

The MOT history is free, and it is the most honest document about the car

Every MOT test in Great Britain is recorded on gov.uk, including the mileage at each test and every advisory the tester noted. It costs nothing, takes about two minutes, and needs only the registration.

We analysed 18,474 MOT tests to see what actually goes wrong. The answer is remarkably concentrated: nearly half of all failures — 44.9% — are suspension, wheels or tyres. That is more than brakes and lights put together. The single most common reason a car fails is tyre tread below the 1.6mm legal minimum, followed by brake pads worn under 1.5mm.

Read that list again, because there is a pattern in it. The most common failures are consumables. They are cheap, they are predictable, and — crucially — they are the things a seller can leave for you to pay for.

Advisories are next year’s failures

Here is the part almost nobody uses properly.

The same tests produced 19,962 advisories against 5,235 failures — close to four advisories for every failure. An advisory is the tester writing down something that is not yet bad enough to fail, but is heading that way. „Tyre worn close to legal limit” this year is „tread below 1.6mm” next year.

So when you look up a car’s MOT history, do not just check that it passed. Read the advisories on the last test and treat them as a bill you are about to inherit. If the last MOT advised worn pads and a tyre near the limit, you are buying pads and tyres — and that belongs in your offer, not in your first year of ownership.

One inversion is worth knowing. Lights are 14.4% of failures but under 3% of advisories, because a bulb either works or it doesn’t. There is no gradual warning. A two-minute walk around the car with the lights on catches something the paperwork never will.

The mileage trail is the fraud check

Because the odometer reading is recorded at every MOT, the mileage history is also the cheapest fraud check available to you. The number should only ever go up.

In our data, about 1 in 70 cars had a reading that went backwards by more than 1,000 miles between tests. Where it happened, a median of nearly 20,000 miles had disappeared from the record.

Not every backwards reading is fraud — a replaced instrument cluster or a keying error at the test centre produces the same pattern, which is why the honest word is „discrepancy” rather than „clocking.” But every single one is worth a direct question before you hand over any money.

The five-minute version

If you do nothing else before viewing a car:

1. Look up the MOT history on gov.uk using the registration. It is free.
2. Check the mileage only ever increases. A drop is a stop sign.
3. Read the last set of advisories and price them into your offer.
4. Ask what the badge implies. No digital service record on a premium German car is a red flag; on a Ford it is ordinary.
5. Walk round with the lights on. Bulbs give no warning.

None of this requires expertise, money, or the seller’s cooperation. It requires about five minutes and a registration number — and it is the difference between buying a car and buying somebody else’s deferred maintenance.

George Arnold is the founder of FindServiceHistory, which retrieves official manufacturer service records and MOT history for UK vehicles. The full dataset behind this article is published in the company’s UK Service History Coverage Report, and its free MOT history checker is open to anyone.