MOTCost: 9,996 pages, one month live, what the data already shows
We built 9,996 car pages from 2024 official UK MOT records. We went live. Here is what the first month showed us, including several things we did not predict.
This is not a celebration piece. One month is not long enough to draw traffic conclusions, and the data is still fresh enough that outliers dominate patterns. What it is useful for is honest early signal — the things readers are reaching for before we have fully built the editorial layer around them, the questions that keep recurring in the feedback channel, and the parts of the dataset that surprised us when we stopped looking at it analytically and started watching real people navigate it.
What We Built and Why
The premise behind MOTCost is simple enough: 2024 UK MOT data is publicly available through official records, but it is not particularly useful in raw form. It requires aggregation, structuring, and a reader interface before a person buying a used Ford Focus can do anything practical with it.
We took 64.2 million tracked tests, organised them by model slug, built pass rate calculations and failure category summaries, and published one page per tracked model variant. That produced 9,996 pages at launch. Not 10,000 — some slugs had data quality issues we excluded rather than publish with caveats. 9,996 is the clean figure.
The Annual Reliability Report established the 2024 baseline across those tests: a 77.50% fleet pass rate, with a 13-percentage-point spread between the strongest and weakest major brands. The car pages index that baseline against specific models and trims.
What we expected: high-volume, well-known cars would attract most of the early traffic. Used car buyers researching specific makes and models before purchase would be the core reader. The failure category data would be secondary interest to most people, who would come for pass rate and leave.
What Surprised Us
Trim-level searches dominate
We expected readers to search by model — Focus, Golf, Fiesta — and then click through to the model page. The early session data shows something different: a significant proportion of readers are arriving with a specific trim already in mind, or searching for trim-specific comparisons (Focus ST versus Focus Titanium, Golf GTI versus Golf Life).
This matters because the trim-level data is where the MOTCost dataset is most differentiated from anything else publicly available. The official records are granular enough to separate Ford Focus ST from Ford Focus Titanium Edition. The mileage profiles, failure compositions, and pass rates differ between those two cars in ways that matter if you are buying one. Readers are finding this faster than we expected.
The Fiesta engagement is disproportionate
The Ford Fiesta is the highest-volume single model in the 2024 dataset: 2,488,311 tracked tests. We knew this. What we did not expect was how strongly the Fiesta page would anchor early reader sessions — not just as a destination, but as a reference point that people return to when comparing other cars.
The pattern seems to be: reader arrives researching a specific car they are considering buying, checks the MOTCost page, then checks the Fiesta as a benchmark. The Fiesta’s pass rate of 78.6% at fleet level — close to the 77.5% average — makes it a useful reference. High volume, well-understood failure profile, mainstream ownership. It is the closest thing the dataset has to a control car.
Readers are more interested in failure reasons than pass rate
We built the pass rate figure prominently. We buried the failure reason data below a fold. Early session recordings show a clear pattern: readers who engage more than 90 seconds almost always scroll to the failure category section. They want to know not just that a car fails at a certain rate but what it fails on.
This makes sense. The pass rate tells you how often cars like yours need repair before passing. The failure category tells you what is likely to need repairing. For someone who has mechanical sympathy — who does their own servicing, or who knows a good independent garage — the failure composition is more actionable than the headline figure.
We are restructuring the page layout to surface failure categories earlier. The data was always there. The reader hierarchy was wrong.
The advisory data is the most-asked-about thing we do not yet publish
We receive feedback through the site. In the first month, the question that appeared most frequently, in various forms, was: where are the advisories?
Advisories are the MOT’s early-warning system. A pass with an advisory means the car passed but has a component that is close to failure — not there yet, but worth watching. For used car buyers, advisories are often more informative than a failure and fix, because they tell you what is coming.
The official UK records include advisory data. We have it. We have not yet built the presentation layer for it, because we prioritised getting clean pass rate and failure data live before layering additional complexity. That decision was right for launch; the reader response makes clear it needs to accelerate.
The Numbers That Emerged
9,996 car pages. This is the model-variant count — not unique models, but unique tracked slugs, each with their own pass rate, mileage profile, and failure composition. A Golf Life TSI and a Golf GTI TSI are separate pages.
64.2 million tests. The volume of 2024 official UK MOT records tracked in the dataset. This is the total across all 9,996 slugs, which means the same physical car is counted once per test year — a car that was tested in 2023 and 2024 appears in both years’ data but only the 2024 figure is in scope here.
77.50% fleet pass rate. Forty-nine million seven hundred thousand passes from sixty-four million tests. The number that everything else is measured against.
13pp gap between top and bottom major brand. BMW at 83.8%, Renault at 70.8%. The gap is structural — it reflects fleet age, ownership pattern, and maintenance culture as much as engineering quality. It has not closed meaningfully in several years of available data.
The most surprising single data point: the Golf R’s 92.93% pass rate on 5,190 tests, with an average test mileage of 24,422. Not surprising that a well-maintained, relatively young performance car passes at high rates. Surprising how clearly the mileage composition explains what looks like an anomaly — the Golf R is simply being tested much earlier in its life than the base Golf fleet, which averages 94,044 miles at test. Same car platform, ten times fewer miles.
What Readers Are Asking
Three question themes dominate the first month of reader contact.
“Why does my specific car have a different pass rate to what the page says?” — Individual car history via the official number plate lookup is separate from the aggregate model data. The MOTCost pages show what all cars of a given type do in aggregate; your specific car’s individual test history is available through the official public record. We have been clearer about this distinction in recent page updates.
“How recent is this data?” — The dataset is 2024 official UK records, published in early 2026. The pass rates reflect the 2024 test year. We timestamp this prominently but the question recurs because people want to know if last month’s test would be in the figures. It would not.
“Why does the Golf have a 78.6% pass rate but the Golf Life TSI has 91.7%?” — This is the trim-mileage question again. The base Golf fleet includes decades of older cars at high mileage. The Golf Life TSI is a newer trim with a younger age distribution. The aggregate pass rate is a mileage-weighted reality, not a trim quality statement. We have added explanatory text to model pages where this gap is large.
What Comes Next
The advisory data publication is the clearest priority. Beyond that, the reader behaviour in the first month has confirmed a few things about the editorial roadmap.
The failure category data needs more prominence on individual pages. Readers want to know what breaks, not just how often a car fails.
The trim comparison content is worth building out explicitly. Hot hatches versus base models, diesel versus petrol equivalents, annual registration cohort differences — these are the comparisons readers reach for organically.
The quarterly snapshot cadence — starting with the Q2 2026 report — gives us a structured way to surface fleet-level movement between annual data cuts. The annual dataset does not update on a quarterly basis; what we are tracking quarterly is our own analytical refinements and any supplementary data that becomes available.
The first month showed us that the core data is what readers came for, and the editorial layer is what makes them stay. The data was always the product. One month in, it is clearer than it was at launch that the editorial work — the context, the comparisons, the explanation of what the numbers mean for someone standing in a car dealer forecourt — is where MOTCost earns the read.
Dataset: 2024 official UK MOT records. 64,183,742 tracked tests across 9,996 model slugs. Fleet pass rate 77.50%. All figures re-derived from source data files — see dataClaims block for verification paths.