Every robot vacuum gets a score from 0 to 100% built from five factors, which maps onto the 5-point rating shown on the site (a 0% score would be 1.5 stars; a perfect score, 5). Each factor is scored against fixed reference points, not against the other products in the list, so adding a new model never moves another model’s rating. Robots that only vacuum, robots that vacuum and mop, and robots with a self-cleaning dock are ranked separately.
| Factor | Measures | Weight |
|---|---|---|
| Cleaning | measured pickup from carpet | 30% |
| Value | current price | 25% |
| Dock | empties, washes, dries, refills | 20% |
| Navigation | LiDAR, cameras, 3D obstacle sensing | 15% |
| Mopping | mop type, lift, edge reach | 10% |
A robot that doesn’t mop isn’t scored on mopping. Its other four weights are scaled up in proportion to fill the gap, so cleaning counts for a third of its score.
The factors
Cleaning comes from independent measurements, not from the maker’s claims. We use Vacuum Wars, which buys robots at retail and runs the same tests on each:
| Test | What it measures | Scoring | Share |
|---|---|---|---|
| Carpet deep clean | Share of sand worked into medium-pile carpet that the robot lifts | 70% scores zero, 95% full marks | 60% |
| Pet hair | Share of flattened 2.5 in pet hair lifted from carpet in two runs | 50% scores zero, 100% full marks | 25% |
| Hair tangle | 7 in hair left wrapped on the brushes after two passes on hard floor | 0.6 or more scores zero, none full marks | 15% |
Embedded sand is the hardest dirt for a robot to lift, so it separates strong cleaners from weak ones better than surface debris, which nearly every robot picks up. Without the carpet test there’s no measured score.
A robot that hasn’t been tested yet is estimated from the brand’s tested robots, pulled toward the median of all 63 tested robots. With one or two brand tests the estimate stays close to the median; with many it follows the brand. The factor is marked estimated until a test arrives.
Value is the current lowest checked price, on a sliding scale per type, with each halving of the price worth the same:
| Type | Scores zero at | Full marks at |
|---|---|---|
| Vacuum only | $800 | $150 |
| Vacuum and mop | $1,000 | $200 |
| Self-cleaning dock | $1,800 | $350 |
A robot with no current price gets the neutral middle score.
Dock scores how much upkeep the dock takes off your hands, weighting each job by the work it saves:
| Dock feature | Weight |
|---|---|
| Empties the robot’s bin | 40 |
| Washes the mop | 25 |
| Washes with hot water | 10 |
| Dries the mop with warm air | 10 |
| Optional water hookup (auto refill and drain) | 10 |
| Doses detergent | 5 |
The score is the share of the weight a robot’s dock earns. Mop features only count for robots that mop, so a vacuum-only robot with a self-empty dock earns full marks. A feature the maker doesn’t say either way about is left out, and the factor is marked estimated.
Navigation scores what the robot uses to map the home and see obstacles:
| Navigation | Score |
|---|---|
| LiDAR + camera | 100 |
| LiDAR + 3D light | 85 |
| Camera + 3D light | 80 |
| LiDAR | 65 |
| Camera | 55 |
| Gyroscope | 25 |
| Random | 0 |
LiDAR maps rooms quickly and accurately, even in the dark. A front camera recognises cables, socks and pet waste so the robot can steer around them; structured light or 3D sensors measure the shape and distance of obstacles but can’t tell what they are. A camera alone maps by sight and is less sure in dim rooms. Gyroscope robots clean in rows but don’t know where they are, and random robots bounce from wall to wall.
Mopping scores the mop:
| Mop | Score |
|---|---|
| Roller | 100 |
| Dual spinning pads | 80 |
| Single spinning pad | 60 |
| Vibrating pad | 45 |
| Static pad | 15 |
The mop type makes up 70% of the factor. A roller keeps scrubbing with a freshly rinsed surface, and spinning pads press down and scrub; a static pad just wipes. The other 30% is split between carpet handling and reach: a robot that leaves its mop at the dock for carpet, or lifts it 20 mm or more, earns the first 15% (5 mm or less earns none), and a mop that swings out to reach along walls and into corners earns the last 15%.
What isn’t scored
Claimed suction (in pascals) is listed but not scored. Makers measure it at the motor in ways they don’t publish, and the claims have climbed past 30,000 Pa without matching gains in measured pickup. Run time, bin and tank sizes, height and noise are on each product page. Vacuum Wars’ area-per-charge figure is shown in the test table but isn’t scored, since most homes are cleaned well within one charge and the robot recharges and resumes when it isn’t.
Missing data
If a figure isn’t published, that factor gets the neutral middle score (50) and the product page marks it as estimated.
Ties and discontinued models
Products are ranked by score within their type. Discontinued models stay listed but rank below every current model and can be hidden with the “Hide discontinued” filter.