Last week I wrote an article about iOS 26’s adoption rates, which, according to StatCounter research, are historically low. But it turns out that the numbers might not be as bad as originally thought.
The thing the report may have missed was a change Apple has made to the way Safari reports version numbers. As Nick Heer points out over on Pixel Envy, Safari on iOS, iPadOS, and visionOS stopped listing the current version of the operating system as of the launch of the 26.0 OSes last year. Or at least, it stopped listing it in the same way.
Heer quotes the User-Agent string for Safari 18.6 on iOS:
Mozilla/5.0 (iPhone; CPU iPhone OS 18_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.6 Mobile/15E148 Safari/604.1
And compares it the UA string for Safari 26.0 on iOS:
Mozilla/5.0 (iPhone; CPU iPhone OS 18_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/26.0 Mobile/15E148 Safari/604.1
As you can see, the correct iOS version is still present later in the string, but the section referring to iPhone OS is now frozen on 18.6. That’s the part which StatCounter could have been using for its data, and that explains why so few iPhone users are showing up as running iOS 26, and so many are still apparently running iOS 18.
(If you’re wondering why Apple would do this, or even suspecting a deliberate attempt to disguise poor adoption rates, it’s worth noting that the company already made this switch on macOS as of Catalina several years ago, as I noticed when I tried to find out if macOS 26.0 had seen a similarly low adoption rate. Maybe that should have been a clue, although it’s not like Tahoe hasn’t taken its share of lumps, too.)
The remaining mystery is why there are any iOS 26 users at all in StatCounter’s data. One possibility, discussed by software engineer Jeff Johnson in a sternly worded blog, is that these are all iPhone owners running third-party browsers: Chrome, Firefox, Dolphin, and so on, rather than Safari. Which makes sense, since the version reporting is specifically a Safari issue… although at that point a total of 15 percent (considering that they have to be both using a third-party browser and running iOS 26) sounds remarkably high. Safari has a huge market share on the iPhone.
Either way, the data is likely flawed, and iOS 26 probably isn’t doing quite as badly as it initially seemed. Its actual performance is hard to gauge and depends on who you ask; Tom Rolfe of TapSmart, for example, kindly emailed me to say his site’s data points to adoption of around 70 percent, which would be perfectly decent, and to suggest that many iPhone owners haven’t even heard of Liquid Glass, let alone allowed it to influence their choice to upgrade or not. You can argue, though, that visitors to a niche Tips & Tricks for iPhone & iPad site would skew toward very high adoption of the latest version of iOS.
Nevertheless, the fact remains that iOS 26 adoption rates are significantly lower than for iOS 18. TelemetryDeck suggests it could be around 55 percent, significantly down from 78 percent the year before, which seems plausible based on the outcry over iOS 26’s divisive changes. Plus, if adoption was the same or higher than for iOS 18, we’d expect Apple to have pushed out data pointing to this by now.
Whatever the truth, we won’t know for sure until Apple chooses to let us. Expect discussion of this topic at the next Apple event, and watch out for any peculiar framing in the way the company discloses the numbers.
link