How many miles does the average van do?

The average, and what it equals

Every MOT records the odometer reading at the moment of the test. Taken across the whole national record, that gives a median figure for what vehicles have actually covered at each age — not what a fleet contract assumed they would.

Median odometer reading in miles at each vehicle age, weighted by number of tests. Source: PriceMyRepair analysis of the DVSA MOT testing dataset.

The same equivalence, arrived at twice

A van at its first test has covered 53,657 miles. A car does not reach that until it is six years old, at 56,052. On distance travelled, three years of van life equals six years of car life.

Separately, and using a completely different measure, we found that 22.40% of vans fail that first test against 22.50% of cars at six years. Two independent measures — how far the vehicle has gone and whether it passed — produce the same conversion between van years and car years.

And the ratio falls the whole way

At three years a van has covered 1.69 times as far as a car. By ten years it is 1.39, and by seventeen it is 1.27. The absolute gap stays roughly constant at around thirty thousand miles from about the eighth year onwards, but as a proportion it shrinks continuously.

That shape is worth holding onto, because it is the same shape as the failure gap on our companion page: widest at the first test, narrowing every year afterwards. Whether one causes the other is the subject of section four, and the answer is less obvious than it looks.

Mileage by age, vans against cars

Both lines climb steeply and then flatten. The van line starts well above the car line and stays above it for the whole of a working lifetime, but the distance between them stops growing after about the eighth year.

The first eight years do most of the work

Between three and eight years old, a van adds roughly fifty thousand miles while a car adds around thirty-nine thousand. After that the two curves rise at almost the same rate, and the absolute difference settles at about thirty thousand miles for the rest of the vehicle’s life.

A van does not out-travel a car steadily throughout its life. It out-travels it heavily and early, then falls into much the same pattern.

What flattening actually means

A van reaching its eighth test has covered around 104,000 miles. By its fifteenth it is at 130,000 — twenty-six thousand more across seven years, or under four thousand a year. That is a fraction of the eighteen thousand a year it was covering when new.

Which vans cover the most ground

At the first test, every vehicle is the same age, so the odometer reading is a clean measure of how hard the model is worked. The range across thirty-two models is nearly four to one.

Median odometer reading at the first test, taken at three years old. Source values are rounded to the nearest five thousand miles in the underlying dataset, so these figures should be read as bands rather than exact totals. Source: PriceMyRepair analysis of the DVSA MOT testing dataset.

The Sprinter is in a category of its own

At 95,000 miles by its first test, the Sprinter has covered nearly twice the van average and three times what a car of the same age has done. The Crafter follows at 75,000. Nothing else on the list is above 65,000.

These are the vehicles doing long-distance haulage rather than urban delivery, and the odometer separates them from the rest of the market far more sharply than any specification sheet would.

At the other end, a van that barely moves

The Ducato records 25,000 miles at three years, less than a car of the same age. That looks wrong until you consider what most Ducatos in Britain are: the base vehicle for motorhome conversions, which cover a few thousand leisure miles a year and spend the winter parked.

The Doblo and the Musso sit at 35,000, both below the car average. A commercial registration does not imply commercial use, and on this measure roughly a tenth of the models on the list are being used less than a typical family car.

Does mileage explain the failures?

Vans cover more ground than cars and vans fail their MOT more often than cars. The obvious conclusion is that one causes the other. Across thirty-two models, that conclusion does not hold.

Fifteen models, identical mileage, eleven points of difference

Fifteen of the thirty-two models record the same 45,000 miles at their first test. Their failure rates run from 13.94% to 25.22% — a spread of 11.28 percentage points between vehicles that have covered exactly the same distance at exactly the same age.

That single column of the chart does more to settle the question than any correlation coefficient. Whatever separates the Amarok from the Combo, it is not how far they have been driven.

The two ends of the chart run the wrong way

The Sprinter covers 95,000 miles, more than any other model, and fails 22.10% of the time. The Primastar covers 55,000 and fails 27.91%. The hardest-worked van on British roads passes its first test more often than one doing barely half the distance.

Across all thirty-two models the correlation between mileage and failure rate is 0.333. That is not nothing, but it is far weaker than the relationship most people would assume, and it leaves the great majority of the difference between models unexplained.

What that something else might be

The candidates are all things the MOT record does not hold: who owns the vehicle, whether it sits on a maintenance contract, what it carries, whether it works in a city or on motorways, and how it is treated by whoever drives it. A Sprinter on a long-distance haulage contract and a small panel van doing urban multi-drop can cover similar annual distances under conditions that have almost nothing in common.

Anyone holding fleet maintenance records alongside registration numbers could test this properly in an afternoon. Until someone does, the honest position is that mileage explains the fleet-level pattern and almost none of the model-level one.

Almost every van is a diesel

The passenger car market has moved. The van market, on the evidence of every test carried out in the record, has barely started to.

All van tests on vehicles aged three to twenty. Share is of tests, not of vehicles registered, so it reflects the fleet actually on the road across the whole period rather than current sales. Source: PriceMyRepair analysis of the DVSA MOT testing dataset.

Ninety-seven per cent, and the alternative is petrol

Diesel accounts for 97.39% of every van test in the record. The remaining fraction is almost entirely petrol, at 2.48%, and that share sits mainly at the small end of the market where a car-derived van shares its engine with a hatchback.

LPG and electric together account for roughly one test in a thousand. Whatever is happening in new van registrations, the fleet presenting itself for testing is a diesel fleet with a rounding error attached.

Fuel makes no visible difference to the result

Diesel vans fail 29.89% of the time and petrol vans 29.75% — a difference of fourteen hundredths of a percentage point across sixty-two million tests. On this measure the two are indistinguishable.

That is worth stating plainly because the categories a van fails on are not evenly split between fuel-related and mechanical items. Emissions account for under four per cent of van failures, so the parts of the test where a diesel and a petrol engine differ most are also the parts that decide fewest outcomes.

What this means for the mileage figures

Every mileage number elsewhere on this page is, in effect, a diesel number. The 53,657 miles at first test, the ratio against cars, the model rankings — all describe a fleet that is 97% one fuel.

That makes the comparison against cars slightly uneven, because the car fleet is mixed. It does not undermine the distance figures, which are recorded the same way regardless of what is under the bonnet, but it is worth knowing when reading anything on this page that touches on running costs or emissions.

What this means when buying

The figures on this page are most useful at the point of purchase, because that is the moment when the difference between a van’s age and a van’s mileage decides what you are actually paying for.

Where the odometer is worth checking carefully

Every MOT records a reading, so a vehicle’s history is a sequence of them stretching back to its first test. That sequence is public, and gaps or reversals in it are visible to anyone who looks. A van showing 40,000 miles after four tests that each added 18,000 is telling you something the seller may not be.

You can pull the full sequence for any registration with our mileage checker, or see it alongside advisories and failures in a complete history check. Neither requires signing up.

What high mileage does change

It does not predict the MOT result well, but it does predict the service schedule. A van covering 18,000 miles a year reaches service intervals roughly twice as often as a car doing 9,000, and reaches mileage-based items such as cambelt replacement years earlier than the calendar would suggest. Intervals and prices are set out in the service cost guide, and test fees in the MOT cost guide.

It also affects what the vehicle is worth. Our guide to valuation covers how mileage and test history interact, and every cost guide on the site is indexed under repair costs. If you want to know what happens to the advisories a high-mileage van collects, we covered that separately in the van advisory index.

We earn a commission if you book through this link, at no cost to you. It does not affect which garages you are shown or what they charge. Our funding page explains how the site pays for itself. You can also check any registration free, with no sign-up, or read what the MOT actually inspects.

Methodology and citation

One public dataset, one measure, and seven things this analysis cannot do. The first two limitations change how every figure on this page should be read, so they are stated before anything else.

The dataset

The DVSA publishes the complete record of MOT tests as open data. This analysis uses the most recent full snapshot: 829,654,015 tests across 137,840,356 vehicles. Every test records the odometer reading taken at the moment of inspection.

We use tests carried out by DVSA only, excluding those recorded by the DVA in Northern Ireland and by the DVLA, and only outcomes recorded as a pass or a fail. Vehicles with an implausible recorded age are excluded.

The measure

For each combination of model and vehicle age we take the median odometer reading, then combine those medians across models weighting each by the number of tests behind it. Vans are identified by model: thirty-two commercial models classified from the vehicle record, with everything else forming the comparison group described here as cars.

Every comparison is drawn at the same vehicle age, because age determines mileage so strongly that comparing two fleets of different average age would show nothing else.

What this analysis cannot do

On the odometer problem specifically

Because readings are unverified, aggregate figures on this page could in principle be pulled downwards by tampering. Two things limit how much that matters here.

The first is scale: distortion would have to be widespread and systematic to move a median calculated across tens of millions of tests. The second is that the same weakness applies equally to the comparison group, so a like-for-like comparison between vans and cars is less affected than either figure taken alone.

Where tampering does show clearly is at the level of an individual vehicle, because the record holds every reading in sequence and a reversal is visible to anyone who looks. That is a feature of the data worth using rather than a flaw to apologise for.

Free to cite

Every figure here may be reproduced in any publication with attribution. No permission is needed and nothing sits behind a form. If a different cut would be more useful — by region, by fuel, by body type, by model — we will produce it on request.

We are not affiliated with the DVSA, the DVLA or any garage group. The analysis, the classifications and any errors in them are ours alone. Corrections are welcome and will be published on this page rather than made silently.