Android fragmentation not so bad says Localytics


Credit: Horia Varian

Android(s goog) fragmentation is less of a challenge than it was two years ago, and developers shouldn’t be as concerned about it says Localytics. The mobile app analytics firm will release a study later on Wednesday showing data to support this idea, suggesting that OS version, screen size and display resolution are now fairly common on most Android devices.

Some supporting datapoints from the findings, which measures device attributes from apps that use the Localytics platform:

  • At 73 percent, nearly three-quarters of all Android devices are running a variant of Android 2.3, also known as Gingerbread.
  • Another 23 percent run Android 2.2, or Froyo, bringing the total percentage of Android devices running these two similar versions up to 96 percent.
  • The majority of Android devices are using either 4- or 4.3-inch screens, accounting for 61 percent of all devices.
  • The 800 x 480 resolution is still fairly standard, accounting for 62 percent of Androids. Surprising to me is that only six percent run the higher 960 x 540 resolution that gained support nearly a year ago.
  • Tablets currently experience less fragmentation, with 74 percent being 7-inch slates at 1024 x 600 resolution, while 24 percent are 10-inch tablets with 1280 x 800 screens.
  • Most of these tablets (71 percent) run on Gingerbread, Android’s phone platform. That’s likely due to sales of the Galaxy Tab, Kindle Fire(s amzn) and Nook Tablet/Color (s bks).

Google’s own dashboard numbers, last updated on Jan. 3, support Localytics’ findings:

I’m in general agreement with the data as well. Although there are hundreds of Android handsets out there, developers using the Gingerbread APIs and supporting 800 x 480 displays are likely targeting the vast majority of currently available Android handsets, not to mention most of the tablets, too. Google has added zoom and stretch functionality in Android 3.2  to assist with supporting different screen sizes as well.

At this point, given that Android fragmentation appeared out of control at one time, I think the current situation is the best Android developers could hope for. The problem isn’t gone, but there are tools to work around it — supporting multiple screen sizes and display densities, for example — and nearly all phones are still arriving with Gingerbread at this point.

I still believe Android 4.0 is Google’s best effort to combat the fragmentation issue. Having used that software both on a phone and now on a tablet, it’s definitely more of a unified experience, although there are still some inconsistencies. Regardless, Android 4.0 is a “fresh start” for the platform on both tablets and smartphones; getting handset makers to adopt it sooner, rather than later, should be a key Google initiative to help both consumers and developers.


Jeff Kibuule

It took a year to get more than 50% of people on the latest version of Android, and I’m sure that only happened because people bought phones that already had 2.3. Apple can do the same in a month and Microsoft can in 2.

Fragmentation may be an issue that a lot of people downplay, but as a developer, it would be stupid of me to build an app that specifically takes advantage of ICS because it won’t be used by most people for almost a YEAR. =/

Kevin C. Tofel

Jeff, I understand but I don’t think it’s as dire a situation as you portray; especially in the last sentence. You’re right, many people may not use ICS for a year. But just today, Mint launched its Android tablet app. It doesn’t use any of the new APIs found in Android ICS and works perfectly fine on the HC and ICS tablets I have in hand. No difference.

Granted, if a developer plans to use some specific APIs or UI elements found only in ICS, he or she is going to see a very limited audience for the app for some time. So I agree that if you “build an app that specifically takes advantage of ICS” it’s a problem. My point is: it’s less of a problem now than it was a year or two ago and as I said, about the best Google could hope for at this point, given prior actions.


Android is most likely always going to be on a slow upgrade process–the OEMs & carriers don’t have much incentive to upgrade asap and most (probably 99%) users don’t seem to care–they probably don’t even know what version they are currently on anyways. And the “informed” user will buy a Nexus lol

Android “fragmentation” affects the developers most, for others it really is a minor inconvenience, usually. Therefore, devs will most likely focus 6 to 12 months behind the current release–writing for the latest version of Android most likely will spell trouble.

Everything has it’s flaws–ios gets “upgrades” asap but you won’t get the features unless you buy the latest & greatest, and not every app will run on the older hardware especially those focused on the latest & greatest version of ios & hardware, and we have yet to see how long MS will support handsets not to mention the open/customizable factor of Android Vs the others.


Reality works over a time arrow, delusions work with closed circles. Means we can already see the next UI advances will be conversational dialogs, while Android is still working out the kinks in strict select dialog.
How many of the coming to market with 2.X right now will be on 4.X and when? How many will be able to support a conversational dialog, after Google is forced to follow suit?

Android Developer

Comments are closed.