Why “20% better” is not “20% less fuel”
Fuel efficiency and fuel consumption are reciprocals, not the same scale. A 20% gain in kilometres per litre is a 16.7% cut in fuel used. A 43% gain is a 30% cut. The bigger the improvement, the wider the gap — and a supplier quoting in km/L always shows the larger number.
This is not a trick anyone needs to be playing deliberately. It is arithmetic that catches people out, and it caught us out: our own internal analysis compared km/L gains against a consumption-reduction claim for several weeks before we spotted it. This page is the correction, written down.
The same result, two numbers
Efficiency and consumption are inverse. Convert with fuel cut = 1 − 1 ÷ (1 + gain), and the two diverge as soon as the numbers get interesting.
| Improvement in km/L | Reduction in fuel used | Difference |
|---|---|---|
| +5% | −4.8% | 0.2 points |
| +10% | −9.1% | 0.9 points |
| +15% | −13.0% | 2.0 points |
| +20% | −16.7% | 3.3 points |
| +30% | −23.1% | 6.9 points |
| +43% | −30.1% | 12.9 points |
| +75% | −42.9% | 32.1 points |
Below about 5% the two are close enough that the distinction rarely matters. Above 20% it matters a lot, and it always favours the km/L figure. That is worth knowing before comparing two suppliers who happen to have chosen different units.
Three of our own results, both ways
Every figure here is from a trial published on this site. Read the two columns against each other.
A taxi at +43.2% km/L
One Toyota Prius, second measurement interval. Expressed as fuel: 30.2% less. And in the trial as a whole — 13,257 km over 106 days — the aggregate is +28.3% km/L, which is 22.1% less fuel. Two legitimate numbers about the same vehicle, thirteen points apart.
A truck at +13.1% km/L
Toyama Transport, aggregate across the whole 20,422 km trial. As fuel: 11.6% less. This is the honest way to describe that trial, and it is the figure we quote.
A refuse fleet at −18.2% fuel
The Waste Services trial was measured in litres per 100 km from the start, so no conversion is needed. Expressed the other way it would read as +22.2% efficiency — the same result, a number four points larger.
Four places the confusion does real damage
Comparing two suppliers
If one quotes km/L and the other litres per 100 km, the first looks better on identical performance. Ask both to restate in the same unit before comparing anything.
Budgeting
Fuel spend follows consumption, not efficiency. Applying a 20% efficiency gain to a fuel budget overstates the saving by about a sixth — on a large fleet that is a material planning error.
CO₂ reporting
Emissions scale with fuel burned. A CO₂ figure derived from an efficiency percentage rather than a consumption percentage will be too generous, and it will be too generous in a document an auditor may read.
Reading Japanese fleet data
Japanese operators record km/L almost universally, European ones litres per 100 km. Any dataset that crosses between the two — ours does — has to be converted before the numbers can be pooled. Most of the large percentages circulating about this product category originate as km/L.