Third-party data on iOS 9 disputes Apple's adoption claims
But data from Mixpanel, which sells an analytics platform to mobile app developers for tracking usage and user engagement, did not support Apple's claim.
Although iOS 9 had accumulated a 39.7% share of all iOS editions as of 4 p.m. PT Monday -- and was several percentage points ahead of last year's iOS 8 at the same point in its post-release timeline -- the newest version remained far behind iOS 7, the 2013 upgrade that has been the adoption benchmark for Apple.
At the identical time mark after iOS 7's debut, Mixpanel had pegged it with a 59.6% share of all editions, refuting Apple's "fastest ever" claim.
The discrepancy between the numbers do not make either Apple's or Mixpanel's necessarily wrong, said Jan Dawson, chief analyst at Jackdaw Research. Each takes a dip into different iOS measurement pools.
Apple mines visits to the App Store for its number, aiming to give developers an idea which versions customers have who are looking for new or updated apps. Mixpanel measures a subset of apps that users have already acquired. Neither are all inclusive.
"Both are proxies for the installed base," said Dawson in an interview.
Dawson analyzed the Mixpanel and Apple adoption numbers and concluded that they were roughly equivalent -- within a few percentage points -- for much of the year. But during September 2014 and September 2015, when Apple issued iOS 8 and 9, respectively, the data sets diverged.
"In both cases, Mixpanel's adoption rate for the new version of iOS was far lower than Apple's, in contrast to the usual pattern during the year," Dawson wrote in an analysis on his blog Monday.
Dawson posited several reasons why that may be so, but the one most Occam's Razor-esque was that consumers who had just upgraded to the newest version were more likely to hit the App Store, eager to find new apps or update existing apps that take advantage of the latest improvements. That tendency could, Dawson said, inflate Apple's take on iOS share.
"Last year, people may have been looking for keyboards [for iOS 8], this year for ad blockers," said Dawson, citing examples of new app categories created by each upgrade.
But iOS 7 didn't fit the pattern Dawson outlined.
In 2013, Mixpanel's adoption rate for that version was significantly higher than Apple's, even as late as a month after iOS 7's release. On Oct. 22, 2013, when CEO Tim Cook Apple touted iOS 7 as having a 64% share, Mixpanel pegged iOS 7 at nearly 79%. (iOS 7, Mixpanel judged, had made 64% by Sept. 29, or more than three weeks earlier.)
Talk about adoption rates and iOS shares is more than academic, of interest only to developers, and more than just bragging rights about which operating system customers grab quickest.
In a separate post Monday (subscription required), Dawson pointed out that for iOS 6 through iOS 8, each succeeding version topped out at a lower level: 2012's iOS 6, for example, reached 94% by Apple's accounting, while iOS 7 and 8 peaked at 92% and 87%, respectively. The adoption rate slump, Dawson added, could be due to any number of reasons, including user apathy, larger-sized upgrades, older devices falling off the list and device owners deliberately choosing to stick with the tried-and-true rather than risk problems.
OS fragmentation has plagued Android devices for years -- the latest version, November 2014's Lollipop, currently has a 21% share, less than iOS 9 accumulated in days -- a trait Apple has at times used to knocked its mobile rival. But iOS suffers from fragmentation as well, albeit one much less substantial.
Thus far, iOS 9 appears to running counter to the fragmentation trend, as it's leading last year's iOS 8 in the preliminary numbers, Dawson said.
And iOS, even at lower peaks, remains the standard other operating systems, even Apple's also-free OS X, must be measured by. "Anyone can get it, you get prompted to upgrade, and after the first 24 hours, it's usually a smooth process," said Dawson when asked why iOS led other OSes in uptake. "These are devices you use daily," Dawson added, unlike a personal computer, Mac or not. "And you get a new phone, to some extent [with each upgrade]."