Further examination could explain why iOS 26’s adoption rates are so low
Last week, I wrote an article about iOS 26 adoption rates, which StatCounter research shows are historically low. But it turns out the numbers may not be as bad as initially thought.
What the report may have missed was a change Apple made to how Safari reports version numbers. As Nick Heer points out at Pixel Envy, Safari on iOS, iPadOS, and visionOS has stopped listing the current OS version since OS 26.0 launched last year. Or at least it stopped listing it the same way.
Heer cites the User-Agent channel for Safari 18.6 on iOS:
Mozilla/5.0 (iPhone; processor iPhoneOS 18_6 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.6 Mobile/15E148 Safari/604.1
And compares it to the UA channel for Safari 26.0 on iOS:
Mozilla/5.0 (iPhone; processor iPhoneOS 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 version of iOS is still present further down the chain, but the section referring to iPhone OS is now fixed at 18.6. This is the part that StatCounter could have used for its data, and it explains why so few iPhone users appear to be running iOS 26, and so many are still using iOS 18.
(If you’re wondering why Apple would do this, or even suspect a deliberate attempt to cover up low adoption rates, it’s worth noting that the company has already made this change on macOS since Catalina several years ago, as I noticed when I tried to find out if macOS 26.0 had seen a similarly low adoption rate. Perhaps that should have been a clue, although it’s not like Tahoe hadn’t taken its share either of pieces.)
The remaining mystery is why there are 26 iOS users in StatCounter’s data. One possibility, raised by software engineer Jeff Johnson in a harshly-worded blog post, is that these are all iPhone owners running third-party browsers: Chrome, Firefox, Dolphin, etc., rather than Safari. Which makes sense, since version reporting is specifically a Safari issue… although at this point a total of 15% (considering they have to both use a third-party browser and run iOS 26) seems remarkable. high. Safari has a huge market share on the iPhone.
Either way, the data is probably wrong and iOS 26 probably isn’t performing as badly as it initially appeared. Its actual performance is difficult to assess and depends on who you ask; TapSmart’s Tom Rolfe, for example, kindly emailed me to say that his site’s data indicates about 70% adoption, which would be absolutely correct, and to suggest that many iPhone owners haven’t even heard of Liquid Glass, let alone allowed it to influence their choice of whether or not to upgrade. You could, however, argue that visitors to an iPhone and iPad tips and tricks site would lean toward very high adoption of the latest version of iOS.
Still, the fact remains that adoption rates for iOS 26 are significantly lower than those for iOS 18. TelemetryDeck suggests it could be around 55%, a significant drop from 78% the year before, which seems plausible given the outcry over iOS 26’s controversial changes. Additionally, if adoption was the same or higher than iOS 18, we’d be concerned. We’d expect Apple to have already released data showing this.
Whatever the truth, we won’t know for sure until Apple lets us. Expect a discussion on this topic at the next Apple event, and watch out for any peculiar framework in how the company discloses the numbers.
