That's how much of the Android market share that Marshmallow managed to capture since its release six months ago. And Google's annual I/O developer conference, where the next-generation Android operating system will be unveiled, is only a couple of weeks away.
And interest in new Android releases is weakening. At the same period in the release cycle of Android Lollipop the operating system was on 9 percent of devices.
By comparison, since its release back in September of last year, iOS 9 is now installed on 84 percent of iPhone and iPads.
The problem is that given Android's current update model, no one wants or cares any more about yearly Android releases.
And by no one, I really do mean no one.
End users don't care about it because it means having to buy a new phone - usually a top dollar flagship handset - every year to get it (unless they buy a Nexus, something which usage data suggests not many do).
Hardware makers don't want it because it means having to put in the effort to customize a new platform. It took until March/April before many of the OEMs even started shipping devices with Marshmallow installed.
The carriers don't want it because it's a new OS to support (and to customize it with its own bloatware).
Developers don't want it because a new release just means more fragmentation.
And even Google doesn't want it because every year it has to try to convince developers that all is OK in Androidland, and that it's actually worth them putting in the effort into supporting the new platform despite the fact that it won't hit double-digit usage in a year.
Here's what it boils down to:
- Android users don't care about updates (even security updates)
- The hardware makers only care about selling new phones
- The carriers only care about contracts
- Developers only want to sell apps
- Google only seems to care about new device activations
Those of you paying attention will know that this is not a new problem, and it's clear that Google doesn't have a clue as to how to solve it. Despite claims that it is working closely with the big OEMs to get updates to users in a more timely fashion, that 7.5 percent adoption rate figure clearly shows that it's not having much of an effect.
And bear in mind that here we're only discussing the fragmentation arising from major releases of Android. When you take into consideration brand, device, and Android versions, fragmentation is a much big problem. To see just how bad it is, take a look at this OpenSignal report from August 2015.
Unless Google is going to unveil some new mechanism for updating Android - which would essentially mean taking it out of the hands of the hardware OEMs and carriers - at the upcoming I/O developer conference, then whatever is announced will essentially be an irrelevance since a year from now it won't even have hit double-digit usage share.