How we compile these figures

Every number on this site is derived from published manufacturer specifications by a repeatable process. This page describes that process, including the parts of it that fail.

Where the data comes from

We collect robot mower listings from major retail catalogues and extract each specification the manufacturer publishes — rated lawn area, cutting width, cutting height, boundary system, navigation technology, rated slope, weight and external dimensions. Nothing is measured by us, and nothing is estimated.

Normalising units

Manufacturers state the same quantity in different units, sometimes within one listing. Lawn area appears in acres, fractions of an acre and square feet. Cutting width appears in inches, centimetres and millimetres. Everything is converted to a single internal unit before comparison, and converted back for display.

This step is where the most damaging errors live, because a missing conversion does not look like an error — it looks like a number. The worst case we found in this corpus was fractions: a machine advertised for 1/4 acre was being read as four acres, because the extraction matched the number after the slash. A mower advertised for an eighth of an acre became eight acres, a figure sixty-four times too large that still reads as a plausible rating for a large machine. It affected more than a third of the listings that state a fraction, and every error ran in the same direction. Conversions are now tested against known values rather than assumed.

Cleaning the corpus

Search results for robot mowers contain many things that are not robot mowers. From the raw set we removed:

• 7 machines that are not robot mowers — push mowers, a corded mower, and remote-control and tracked machines that a person drives. The last of those are real products, but they have no boundary system, no coverage rating and no autonomy, so they corrupt every average on this site.
• 2 accessories — weatherproof garages and shelters sold on their own.

Exclusion patterns have to be narrow. An early version matched the word “garage” and removed six genuine mowers sold bundled with one — a mower packaged with its garage is still a mower. The rule now distinguishes a listing that is the accessory from one that merely includes it.

What we keep even though we cannot classify it

15 machines in this corpus publish no navigation technology at all. They are kept: they are real mowers with a real rated area, price band and boundary system, and the finder can place them on every one of those axes. They are simply absent from the category pages, which are organised by the one thing those listings do not say. Dropping a sound product over a single unstated field is a loss a corpus this size cannot afford.

Where physics rules a figure out

A robot mower’s deck adjusts within a range that real machines occupy. Listings that published a cutting height of 457 mm — eighteen inches — were publishing something that is not a cutting height, usually because the specification table carried a maximum where we read a minimum. In those cases the cell is discarded and the rest of the listing kept, because the remaining figures are sound and discarding the whole machine would lose a real mower over one bad cell. The same guard applies to cutting widths below 120 mm, which no mower deck reaches.

The guard is deliberately wide at the top. Large-acreage machines genuinely cut 20 inches, and Amazon’s own specification table states it — an earlier version of this check capped the range at 400 mm and threw away twelve real decks that the source had right.

How baselines are computed

A baseline is the median of a specification within one group. Groups here are navigation technology, lawn-size class, and boundary system — each taken on its own.

That is a departure from how a larger corpus would do it, and the reason is arithmetic. Three technologies by five size classes is fifteen combinations, and this corpus fills only two of them with enough listings for a median to describe anything. Computing the other thirteen anyway would produce numbers with the appearance of precision and none of the substance. So each axis is measured on its own, where every group clears the bar, and no page claims a figure for a combination that was never measured.

Lawn-size classes are: under 1/8 acre, 1/8 – 1/4 acre, 1/4 – 1/2 acre, 1/2 – 1 acre, 1 acre and above.

A group needs at least 8 listings before a baseline is published. Below that the median describes the sample rather than the market and would move if one more listing appeared. 10 of the 10 groups in this corpus clear that bar; the rest are shown as unavailable.

Claim density, and what it is not

Claim density is the one figure on this site that no manufacturer publishes: rated lawn area divided by cutting width, in square feet per centimetre of deck. It is arithmetic on two published numbers, not an estimate.

It is not a quality score and it is not a measure of speed. A machine well above its class median is claiming to cover more lawn with a narrower cut, which it can only do by running longer — so the figure says how hard the rating is working, not how good the mower is. We publish it because the alternative is taking rated areas at face value across machines whose decks differ by a factor of five.

What we will not publish

• Star ratings and review counts. Not published in any form, including paraphrase.
• Exact prices. Prices change daily; a scraped price is misleading by the time it is read.
• Estimated specifications. Where a manufacturer has not published a figure, the page says not stated.
• Test results. We do not test, so we have none.

Known limits

Specification coverage is uneven and this constrains what can be compared. Rated area is published by most listings; a slope rating by under half. Run time, charging behaviour and noise are too sparse across this corpus to support any comparison at all, so none is offered — rather than a comparison built on the fraction of machines that happen to state one.

The corpus reflects one retail catalogue at one point in time. It is not the whole robot mower market, and machines sold only through garden-machinery dealers are absent. Figures were last computed on 2026-08-27.