Anecdotally the only person I know who owns a roomba used to work for iRobot.
If we assumed a half-life of n years, then robot vacuums will only ever approach 20% asymptotically and won't ever reach it.
(We could get over 20% if robot vacuums have a longer lifespan than regular vacuums. I'm tempted to write a simulation over different failure curves but I really have better things to do.)
You shouldn't assume that they will be 20% of sales for every future year.
[1] https://business.highbeam.com/industry-reports/equipment/hou...
It's not perfect. I have a single-floor apartment with no door thresholds, yet it keeps getting stuck. I keep cables and so on neatly organized and hidden, but sometimes I've left a temporary charger cord or something on the floor; this kills the Roomba. You also have to careful with plants -- it ate a whole branch of a plant once that was hanging low near the floor. It has issues with rugs, especially with black areas, but also just getting onto the rug. It also loves to close the door on a room so that it can't get out; I had to buy a couple of door wedges to prevent that, but I don't always remember to use them. Pretty much 50% of the time I come home to find the Roomba dead in a corner.
Its dust compartment is quite small and must be emptied at least every 2-3 days. You also have to take out the little spindle things and remove hair which collects around the axles. Fortunately, the Roomba is really modular in design; you can take it apart completely, and there are replacement parts for absolutely everything. It is a rare product where the designers seem to have rejected the trend of planned obsolescence; I haven't really owned it long enough to be qualified to judge, or taken it apart, but everything I have seen indicates that it was carefully engineered to be repairable and last a long time. (It's probably hackable, too.)
Despite its weaknessss, I am quite happy with it. It keeps my apartment clean and removes the mental burden of thinking about dust. A couple of times a month I take my Dyson to clear up some corners that the Roomba can't get to where dust bunnies pile up. I have a friend who combines it with the cleaning model, but I can't deal with two gadgets that leave their depleted, cable-choked corpses around.
But after I moved to a different apartment, and he got his own place. I really saw how much work that thing was doing. I forgot how often the floors need to be swept to keep it clean. I plan to buy one myself once I move at the end of this month.
I had one about five years ago, and so long as I pressed the "clean" button every day when I left my flat it was good enough that I only manually cleaned every two weeks or so.
My place was ideal though, all on one floor, with no steps, and polished wooden floorboards + thin carpets only.
I expect modern ones are better in terms of suction, battery life, and I know that you can schedule them so that removes the element of having to remember to press the button
You can't have anything on the floor, so I wasn't really able to let it run while I am not at home because I was afraid it would hit stuff. Sometimes it even pushes furniture around or bumps into other stuff and scratches it.
I can vacuum my whole apartment in a lot less time than when using the roomba.
The roomba kept blowing around the dog hair and ignore spots, but kept going over the same spots a thousand times.
I'd say this has had a similar impact on our life as a dishwasher.
Also, newer ones are automated, so you come home every day, empty the canister, and that's your floors cleaned.
Just as you said too, roombas come with an Infrared wall, basically to make a barrier between something. They also have a "hub" you can by, which if you sat between feeding bowls puts a ring around them the roomba wont cross.
The cleaning and coverage is pretty good. My Roomba didn't have massive cleaning power but because it revisits the same spot over and over it got the job done.
It required quite a lot of attention though, both emptying it and removing hair and carpet fluff from it.
However the real reason I got rid of it was that I suspected that the filtration wasn't very good and that it was just sucking up allergens and blowing them into the air as it went.
It's a difficult thing to measure though and I didn't want to take the chance again (in the short term).
I found I got better cleaning (measured anecdotally by how much Ventolin I use) from a high-end handheld vacuum cleaner with HEPA and changing how I clean to include curtains, mattresses and pillows.
Definitely not as cool as having a robot though. I think that Karcher used to make a robot with a base station that it would empty itself into, which is a neat idea. That would cut down on the emptying.
[EDIT] I should add that the handheld vacuum cleaner has its own problems too since it's bagless. Great in principle until you empty it into a cloud of dust. I like seeing how much dirt is picked up but maybe disposable bags are the best solution after all.
I had one of the early 400 series and it was problematic: it required a lot of picking up and it would get entangled on cords, carpet fringes, anything loose. It also would get stuck and the brushes would get jammed up with hair. So while it vacuumed ok, it required so much picking up and fiddling with that we gradually stopped using it.
The new 805 we got at Costco (a bargain too) looks similar overall but everything is different in detail. The new rollers don't get tangled, it doesn't get caught on cords or fringes, it slows down and approaches walls cautiously instead of just crashing into them, and the algorithms for getting unstuck are effective. Now we have it set to run unattended twice a week and the house is cleaner than it has been for years.
I recommend the 800 series (or up I guess) over the 650 that Sweethome liked because it has the new rollers and bin so it picks up better and needs much less hair removal and other fiddling around.
I'd pay serious money for a real cleaning robot though. Not a little disc but something that can navigate stairs, clean higher surfaces. Currently paying humans to do it. They have access to my house, etc.
Once that cleaner robot really exists, imagine the extensions to it. Give it some other sensors and you have a fire alarm and firefighter. Even overall security. Close the doors behind you. Turn off something you forgot.
Rather than put chips in every item and automate everything I'd rather have one big thing to automate my (legacy) house. Easier to upgrade, fix plus I'd have a fully functioning house if the logic fails.
Looking at Japan, they seem to lead in this area. Can't wait.
I bought it at Costco so I'm considering just bringing it back, but I feel kind of guilty after waiting so long.
I made the mistake of buying one for my parents soon after I got mine for Christmas. Theirs eats the low hanging drapes, so it sits in the corner as well. I always wince at it when I go over there for dinner.
Don't do it.
This is a similar story to Fitbit and GoPro, once you've own the market you're after, then what?
I've tried cheaper competitors and they aren't even close to how well Roomba's work. As long as they keep building the best quality robots, I'll keep buying them. If they divert resources to a new market and forget about their main product (kind of like Apple has with Mac), I have no issues in buying a competitors product if it's better.
Then on to lawn mowing. They have figured this out already, just none of them have as much name recognition as the roomba.
That said, the reviews trickling into Amazon kind of deflate those hopes: suction is great, but otherwise it seems to have all the kinds of problems that all the robot vacuums have.
Definitely not something I'd be willing to plunk down a thousand dollars for. Here's to hoping the next version is a marked improvement.
Let's say there's one vacuum for every 20 people (household of 4 or so, 1/2 households who can afford it own one, 1/2 households can afford it), that's 350 millions vacuums.
They've sold 14 millions robot vacuums to date, let's say 25% are broken/not in use, that's 10 millions.
10/350 is 3%. According to some web page I just googled, IRobot has 60-65% market share, so accounting for competitors that's 5% of vacuums.
There is no reasonable interpretation of the public data (and estimates for privately held vacuum mfgs) where 20% becomes true. Perhaps the dataset is their employees and found only 20% own their product? ;)
Can someone explain to me the draw?
If you're happy to vacuum weekly and are already, don't buy one (seriously), but if like me you don't recall the last time you manually vacuumed then buy a $300 Xiaomi Mi and let it do your chore for you.
Plus I've found that they get into places which people ignore (e.g. under the bed, bottom of the closet) and dust still acculates in those places.
It may not have amazing suction or intelligence (though I hear newer ones are smarter...), but it will randomly cross over the same spot a few dozen times during its run, and that seems to make up for it.
For me, this means that my environment can be cleaned daily while I exercise or brush my teeth. Manual, time consuming, but admittedly more effective cleaning can now be done far less frequently.
Same with a Roomba - I basically just quickly check a room to see if there's anything lying on the ground, throw it in there and come back in a hour. Clean and in meantime I can actually do my work or something more fun.
And even worse, just past the warranty period the side brush which rotates to flick dirt into the path of the irobot broke.
I guess I am part of the 20% even though the irobot sits in my cellar doing nothing.
We bought a dyson fluffy (battery powered) recently and that's a much better choice.
I actually find that beneficial. Takes me a couple of minutes and ensure that I clean up more often.
> Worse still it never seemed to learn and would often finish its job leaving huge parts of the floor untouched.
That is on purpose. Different than other robots, Roombas don't learn a floor-plan, but follow some simple, "insect-like" rules. (Travel along if you find a wall, turn around if you hit something, etc.)
It uses the length of stretches that it can drive in a straight line, to get an estimate of the room size, which it will use to adjust the runtime of it's cleaning program.
This behaviour ensures that Roomba adapts well to different rooms. It doesn't clean the entire floor in one go, but over some runs it will eventually reach everything.
And whether the insect like rules are on purpose or not, the fact remains that after 7 days there's still a huge dust ball on my kitchen floor that the irobot continually fails to find.
Oh, and a good thing, this is another person we trust to take care of the house in emergency when we are not here, she lives 500m from our house, we cultivate social relationships at the "village" level. This makes a nice smile and short conversation when we meet here and there.
So, for us, no robots, but for a reason not exactly the same as your reasons. We prefer to trust people.
Imagine a world where everything is automated, you just use stuff which are autonomously delivered to your door and you work from home, you order your pizza with an app, etc.. You are alone.
The other side of the coin is also in security, people by nature make your house less secured (albeit based on your relationship with them, the area, country and so forth).
Not an irobot but something similar. Luckily it was only 120€.
This thing eats everything. Socks. Cloths, papers. Cables, anything on the floor stops, breaks, halts that thing. It really needs some kind of sensor to detect stuff like that before it gets any kind of useful.
I protest! It seems to carefully avoid anything that resembles a crumb, a hair or dust.
Headline should read, "Chap selling super expensive robot vacuum cleaners that take ten times as long as humans, and need just as much supervision, claims they're not irrelevant, really good Christmas present. Honest."
I'd believe that 20% of the world's new robots are vacuum cleaners though.
