Is a robot vacuum actually worth buying?

Robot vacuums have got genuinely good at navigation and genuinely no better at picking dirt out of carpet, because that is a physics problem rather than a software one. A robot has a small motor, a small battery and a bin the size of a mug, so it cannot generate or sustain the airflow an upright does. What it can do is run every day without anyone deciding to, and daily light pickup keeps a floor far cleaner than a thorough clean every other weekend. Judging one against a proper vacuum is the wrong comparison; judging it against the floor not being vacuumed at all is the right one.

What you'll need

• An honest look at your floor types and thresholds
• Somewhere at floor level, against a wall, for a dock
• A count of trailing cables at floor level (optional)

Step by step

1. Check your floors against what a robot is good at.Hard floors and low-pile carpet: genuinely good, especially for dust, crumbs and pet hair. Medium pile: adequate on the surface, nothing below it. Deep or shag pile: it will struggle to drive across it, let alone clean it. Long-pile rugs and rug tassels stop most models entirely.
2. Count the thresholds and the levels.Most models climb about 20mm. A raised threshold strip, a step down to a kitchen or a change of floor level will define the edge of its world. Stairs end it altogether — a robot cleans one floor and has to be carried, which most people stop doing within a month.
3. Look at the floor as an obstacle course, honestly.Phone chargers, lamp flexes, shoelaces, socks, children's small toys, cat bowls, curtain hems that touch the floor. A robot works well in a house that stays tidy at floor level and fails constantly in one that does not. This is the single biggest predictor of whether it gets used.
4. Decide whether you want mapping and no-go zones.Cheaper models bounce randomly and cover a room eventually. Mapping models with lidar clean in rows, know where they have been, resume after charging and let you draw exclusion zones round the cat bowl and the cable nest. For anything beyond one open room, mapping is the feature worth paying for.
5. Be realistic about the mop function.A pad dragged over a floor with a trickle of water is a damp wipe, not mopping. It is genuinely useful for keeping dust film off hard floors between real mops; it will not lift a dried spill or clean a kitchen floor properly, and on many models the pad must be removed before it goes near carpet.
6. Cost the consumables and the dock over five years.Brushes, side brushes, filters and eventually a battery, plus dust bags if you buy a self-emptying dock. It is a real ongoing cost and worth pricing for the specific model before buying, exactly as for any appliance.
7. Consider the self-emptying dock seriously if dust matters.It is the difference between emptying a mug-sized bin every run and dealing with it monthly. In an allergy household it also keeps the dust sealed, which matters more than the convenience.
8. Check the privacy position if the model has a camera.Mapping and camera data go to an app and often to the manufacturer's servers. Decide whether that is acceptable to you before buying, and keep camera models out of bedrooms and bathrooms if it is not.
9. Decide what it is replacing, and be honest about it.If the answer is a weekly vacuum you would otherwise do properly, you will be disappointed. If the answer is a floor that gets vacuumed when someone gets round to it, a robot running daily will leave your floors visibly cleaner than they are now.

Tips

• Run it daily rather than weekly. Its advantage is frequency, and a daily light pass on a floor beats a heavy pass once a fortnight for how the floor actually looks.
• Set it to run while the house is empty. They are not quiet, and the noise is what makes people stop scheduling them.
• Cable-clip everything at floor level before the first run rather than after the first jam. Cables are the most common cause of a robot being abandoned.
• Clean the brushes and the drop sensors weekly. Performance falls off within a fortnight otherwise, and it is usually mistaken for the machine being poor.

Common mistakes

• Buying one as the only vacuum in a carpeted house — It cannot generate the airflow to lift grit out of pile, and its bin holds a fraction of what one room produces. It will keep the surface tidy and leave the carpet dirty.
• Buying a random-bounce model for a multi-room house — Without mapping it cannot cover rooms systematically or resume after charging, so large or complex layouts get partially cleaned at best.
• Expecting the mop function to replace mopping — It is a damp pad with no pressure and very little water. It maintains a clean hard floor; it does not clean a dirty one.

Alternatives

• A cordless stick vacuum instead: Similar money, and for many households more useful — it does stairs, upholstery, the car and the corners a robot never reaches, at the cost of you pushing it.
• Robot plus an existing vacuum: The configuration that actually works. The robot maintains, the real vacuum cleans properly every week or two.
• Cleaning help: For the same annual cost as a mid-range robot plus consumables, some households get more value from a few hours of help a month, which also cleans everything above floor level.

Questions people ask