Every robot vacuum cleaner is put through the same series of tests, so that we can can work out which ones are good and which are bad. We measure quality (vacuum and mop performance) and usability, including navigation and app control.
Our test environment
We test our robot vacuum cleaners in a real environment, complete with chairs and other furniture. We have a combination of hard floors and carpet, so that we can test the cleaners’ performance on both types of surface.
We take before and after images of our dark carpet to show how well the cleaner did. We use a ‘X’ of flour in the middle of the carpet to simulate a real-life spill of dust. This is a hard test, and more dirt than most cleaners would need to pick up, but it lets us spot the good cleaners from the bad.
Hard floor test
Similar to the carpet test above, we also spread X on a hard floor, both in the middle and at the side of the room, so we see how regular performance is and if the robot can clean the edge of a room.
For those robots that can mop, we test on a light floor that has been naturally soiled by being walked on. We take before and after images to see if the robot can deal with tough ground-in stains or just lighter surface dirt.
We measure the sound of each vacuum cleaner by standing above it with a decibel meter. We measure the sound on all performance modes.
As well as testing the quality of cleaning, we measure each robot cleaner to see how easy it is to use. We use the apps thoroughly to see what options are available, testing no-go zones if available.
We also see how well each robot can navigate a room, making note of any times that it gets stuck and where it struggles.
Scoring and verdict
All ratings are scored against the criteria mentioned here. We weight reviews based on price and intended audience. For example, a budget cleaner that has a few navigation issues but generally cleans well can score highly; an expensive cleaner that misses chunks of dirt behind would be downgraded.
If we’ve missed anything that you think we should cover, get in touch with us on Twitter @TrustedReviews or you can email the editor at firstname.lastname@example.org.