According to Statcounter, Windows 11 held a 55.18% market share in October 2025. That share dropped to 53.7% in November and dropped again in December. Now, Windows 11 holds a 50.73% market share.
https://gs.statcounter.com/os-version-market-share/windows/desktop/worldwide
Many are rollback to Windows 10, but Linux is increasing as well.
https://gs.statcounter.com/os-market-share/desktop/worldwide
How do they measure these stats?
They aren’t reaching into my PC so they’re only checking when I [X].
So it isn’t it always just measuring “Os percent from user who [X]”
E.g. Steam only check people with Steam. Slash Dot can only going to measure PCs who go to Slashdot. AOL.com is only checking Boomers.
Statcounter is running on more than a million websites. They track user metadata across these websites.
While this doesn’t give you absolute numbers for everything, it should be enough to notice trends.
Their methodology is on their website.
Agreed, there is no objectively perfect way of measuring this stuff. My point mainly is that the author of the article picked one data point, took it out for context and built an entire lie on that. It’s very much a “look at this snow - so much for global warming” argument. But also, we keep hearing how much Windows is tanking and yet all the metrics we have show it’s actually doing well. Do people like it? No, I don’t think they do. Do I personally want to see Windows crash and burn? Yes, at least in it’s current form. But for all the frustration and anecdotes it doesn’t look like it’s going anywhere, and I don’t think any decision-makers will be convinced that Windows is failing when all the available stats suggest otherwise.