How We Review
Let us be direct about this first
We do not physically test these machines. We do not own a fleet of robot mowers, we do not have a test lawn, and we have not run any of these models across a season of real use. Any site that implies otherwise about an entire category is asking you to take a lot on trust.
What we do is documentary analysis: we gather what the manufacturers publish, normalize it so it can be compared, weigh the factors that decide real purchases, and state plainly where the information runs out. That is a genuinely useful thing to do in a category this opaque - but it is important you know which kind of source you are reading.
What we analyze
1. Official manufacturer information
The manufacturer's own product page, specification sheet, user manual and support documentation. This is our primary source for every figure. We do not take specifications from retailer listings, marketplace descriptions, or other comparison sites, because those are where transcription errors and stale figures multiply.
2. Technical specifications
Rated mowing area, rated maximum slope, navigation technology, cutting width, cutting height range, runtime per charge, charge time, sound level and ingress protection rating. We convert everything into consistent units so models can be compared line by line rather than approximately.
3. Warranty
Length, what it actually covers, whether the battery is covered on the same terms as the body, and whether service is available in the United States or requires shipping the unit abroad. For imported brands this is frequently the most important difference between two otherwise similar models, and it is almost never in the marketing.
4. Navigation
Which positioning technology the model uses, and what that technology needs from a yard to work properly - clear sky view for satellite RTK, adequate light and visual contrast for camera systems, and so on. We treat navigation type as a compatibility question about your property, not as a feature to rank.
5. Terrain and lawn size
How the rated area maps onto real lawns, and what leaves no margin. We flag where a model is close to its ceiling rather than reporting the number without context.
6. Maximum slope
The published gradient, converted and explained, plus the qualifications the manufacturer attaches to it. We note when a rating applies only to driving straight up a slope rather than traversing it.
7. Verifiable owner feedback
Patterns in published owner reports - recurring faults, recurring praise, and recurring complaints about support. We use this to identify themes, never to manufacture a score. We do not aggregate ratings into a number of our own, and we do not publish star ratings of any kind.
8. Price against capability
What a model delivers for its price relative to the nearest alternatives in the same size and slope class. Where a cheaper model covers the same ground within its published limits, we say so.
What we will not do
- Claim to have tested a machine we have not tested.
- Publish a specification we cannot trace to an official source. An empty cell is an honest answer; an invented figure is not.
- Invent ratings, scores, awards or certifications, or emit structured data containing them.
- Let commission rates decide what we recommend, or sell placement on our pages.
- Quote a price without the date we checked it.
How our badges are decided
Labels such as Best for small lawns, Best for slopes and Best value describe the constraint a model handles best within the set we track. They are editorial judgments made from published specifications, not test results, and we name the reasoning next to the badge so you can disagree with it.
Keeping it current
Prices and specifications change. Every model carries the date we last verified it. When a manufacturer revises a specification or a model is discontinued, we update or remove the entry rather than leaving it to age quietly.
Tell us when we are wrong
If a figure here is incorrect, send it to us with the official source through our contact page. We will correct it and re-date the entry.
Method last updated: September 12, 2026.