Guide
Why your car's speedometer reads higher than GPS
Short answer
Your car's speedometer is legally forbidden from under-reading, so manufacturers deliberately set it high. Under UNECE Regulation 39 — the rule across the EU, the UK and much of the world — an indicated speed may never be below the true speed, and may exceed it by up to 10 per cent plus 4 km/h. Most cars sit a few per cent into that allowance, which is why a dashboard showing 100 km/h often means about 95.
GPS is under no such obligation. It reports what it measures, and on a modern receiver with a clear view of the sky it measures well. So the gap you are seeing is not a fault in either instrument: it is the difference between a number designed never to get you in trouble and a number trying to be right.
What the dashboard is actually doing
A car speedometer does not measure speed. It counts how fast something rotates — a wheel, a driveshaft, an ABS sensor ring — and converts that into a speed using a fixed assumption about how far the car travels per rotation.
That assumption is the weak point. It is set at the factory for a new tyre of the specified size, at the specified pressure, at a nominal load. Every one of those changes in normal use. A tyre that has worn from new to the legal limit is a few millimetres smaller in radius, travels slightly less per turn, and therefore reports a slightly higher speed than it did when the car was delivered.
This is precisely why the calibration is biased upward. If the manufacturer aimed for exact, half the cars on the road would be under-reading within a year of tyre wear, and every one of them would fail the regulation.
The rule that forces the bias
UNECE Regulation 39 sets the boundaries in every market that adopts the UNECE framework, which includes the European Union, the United Kingdom, Japan, Australia and many others. The requirement, in plain terms:
| Condition | What it means for you |
|---|---|
| Never below true speed | The dashboard may not show 95 when you are doing 100. Under-reading is not allowed at all. |
| At most 10% + 4 km/h above | At a true 100 km/h the dashboard may legally show up to 114 km/h. Most cars are far more conservative than the ceiling. |
| Typical real-world bias | Around 3 to 7 per cent high on a modern car with reasonably fresh tyres, growing slightly as they wear. |
Notice that the tolerance is one-sided. There is no equivalent rule pushing accuracy the other way, so there is no engineering incentive to be exact — only to be safe. The result is a fleet of cars quietly overstating speed, which is also convenient for everyone involved: the driver who thinks they are doing the limit is usually a little under it.
How GPS gets to a speed
There are two ways to derive speed from a satellite receiver, and they are not equally good.
The naive method is to take two positions a second apart and divide the distance by the time. It works, but it inherits all the noise in each position fix, and at low speeds the noise can be larger than the movement — which is why cheap or badly written apps show you drifting at 4 km/h while parked.
The method modern receivers actually use is Doppler. As you move relative to a satellite, the frequency of its signal shifts, exactly as the pitch of a siren changes as it passes you. The receiver measures that shift across several satellites at once and solves for your velocity directly. It never has to know where you are to know how fast you are going.
This matters for the comparison, because Doppler-derived speed has properties your dashboard does not:
- It is unaffected by tyre size, tyre wear, pressure or wheel slip.
- It does not drift with age or need recalibrating.
- It is measured directly rather than inferred from a proxy.
Where GPS is the weaker instrument
None of which makes it infallible. GPS has failure modes the dashboard simply does not have, and being honest about them is the difference between a useful tool and a false one.
- It needs sky. In a tunnel there is no signal at all. Between tall buildings, signals arrive after bouncing off glass and concrete, and a reflected signal is a wrong signal.
- It is a ground speed. On a steep climb your car travels further along the road surface than it does across the map. The effect is genuinely small — even a punishing 10 per cent gradient accounts for around half a per cent — but it is real, and it always points the same way.
- It is not certified. No phone app is a legal measuring instrument, and neither the app nor your dashboard is a defence if you are stopped.
- The phone matters. Receiver quality varies enormously between devices, and a phone lying flat in a door pocket sees far less sky than one mounted near the windscreen.
Working out your own car's offset
The gap is close to a fixed percentage rather than a fixed number of km/h, which means it grows with speed. Ten minutes on an open road tells you what yours is.
- Pick a flat, straight road with an open view of the sky. Avoid tree tunnels, cuttings and anywhere hemmed in by buildings.
- Mount the phone where it can see out, and give it 30 to 60 seconds to settle before you trust the first reading.
- Hold a steady indicated speed — cruise control is ideal — and let both numbers settle for several seconds. A glance is not a measurement.
- Note the pair. Repeat at 50, 90 and 120 km/h, or 30, 50 and 70 mph, and if you can, run the same stretch in both directions to cancel out wind and gradient.
Divide indicated by true and you have your car's multiplier. Somewhere around 1.03 to 1.07 is entirely normal. Once you know it, the dashboard becomes readable: you are not looking at a broken instrument, you are looking at a consistently offset one.
Do this as a passenger, or with cruise control on an empty road. Reading two instruments and doing arithmetic is not something to take on while also steering.
So which number should you drive to?
For staying legal, the dashboard, because it errs in your favour: if it says 50, your true speed is 50 or a little under. Driving to the GPS number instead means routinely sitting a few per cent above the indicated speed, which is legal only until it is not.
For everything else — how fast you actually went, how far you actually travelled, what your average really was over a two-hour drive — GPS is the better answer, and it is the one worth recording. That is what a trip log is for.
Common questions
Why does my car's speedometer read higher than GPS?
Because regulation forbids it from reading low, so it is calibrated high with margin for tyre wear. Typically 3 to 7 per cent above true speed on a modern car.
Is GPS speed more accurate than a car speedometer?
Under open sky, generally yes — Doppler-derived speed is precise and immune to tyre wear. In tunnels, urban canyons and heavy tree cover it is worse than the dashboard, and it is never a certified instrument.
Which speed do speed cameras use?
Their own, measured with radar, laser or loops in the road. Your dashboard has nothing to do with it, which is exactly why the difference is worth knowing.
Do bigger tyres change my speedometer reading?
Yes. Taller tyres cover more ground per rotation than the calculation assumes, so the dashboard under-reads; worn or smaller tyres make it over-read. Around one per cent across the life of a set is normal.
Can I use a GPS app to check my speedometer error?
Yes, and it is the simplest method available to a private owner. Steady speed, flat road, open sky, several speeds, both directions if you can.
Find your car's real number
Speedometer One shows GPS speed live and records the whole drive, so you can compare at a glance and read the trip properly afterwards. Free on Android.
Disclosure: I am the developer of Speedometer One, the Android app this site is about. Regulatory figures here refer to UNECE Regulation 39 as adopted in the EU and UK; rules differ elsewhere, and your own vehicle's approval documentation is the authority for your car. Nothing here is legal advice, and no app reading is a defence against an enforcement measurement.