We're here with another round of comparisons between Google's platform numbers and our own.
Google has recently released their latest numbers for the Android Platform Dashboards; detailing the breakdown, by platform version, screen size, etc., of users visiting the Google Play Store on their devices. We decided to take that opportunity to revisit our previous post on Corroborating the Android Platform Dashboards with an update as well. For starters, here are Google’s numbers for the collection period ending 5/1/2014:
|2.3.3 – 2.3.7||Gingerbread||10||16.2%|
|4.0.3 – 4.0.4||ICS||15||13.4%|
Double Encore has collected the Android user data from five of our top applications for the same timeframe to compare. As in our previous post, we will try to add some additional context to the overall platform data as well. We won’t be drawing too many parallels between the posts, as the application set between them is not the same.
This data is direct from the Play Store Developer Console, not a third party analytics tool, so our data is coming from the same source. All of these applications have obtained 400,000 or more downloads (most are in the 1,000,000+ bracket), and each is in a separate store category, so we believe this data represents a relevant and valid user sampling. We will be displaying the following metrics:
- User breakdown by country (Where do the users live?)
- User breakdown by manufacturer/device (What hardware do our users own?)
- User breakdown by platform version (Are the users on an updated device?)
The numbers collected and published here for users by country are the “current installs by user” numbers from Google Play. The other metrics are taken from the “current installs by device” data. If you want more information about how Google defines these terms, have a look here. The following data set represents a total of 2.25 Million unique users.
Let’s first have a look at where these users live. This impacts what devices they have access to, so it provides context for the remaining metrics.
The users of our applications are very concentrated in North America, specifically the United States. The top three countries (United States, Canada, Australia) account for 90% of the users.
At Double Encore, we focus our testing efforts around device manufacturer and device family. By and large, devices in the same family from the same manufacturer behave in the same manner, as they often are even running the same system image firmware. A key metric in this is the breakdown of user devices by manufacturer.
Samsung has the majority share by a wide margin here, and over 91% of the most-used devices come from just five manufacturers: Samsung, Motorola, HTC, Google (which includes the Nexus line), and LG. This proportion follows with the country breakdown of users. If our user base skewed more heavily to Europe or Asia, for example, we would probably see manufacturers like Sony or Huawei with a stronger presence in the survey. Let’s zoom in and take a look at some numbers by unique device.
This graph discretely calls out the 25 most popular devices of our aggregated users, which make up almost two-thirds of the active users. The Samsung Galaxy S III is currently in the top spot, with the S4 not far behind; the Galaxy S5 currently holds <1%. Overall, there were 6580 unique device signatures for the aggregated active user base (most with fewer than current 10 installs), with the top 90% of users represented by 128 unique devices types.
Users by Platform
Finally, we want to see what version of Android our users have on their devices. This is, of course, the metric most of us use to determine what versions of the platform to support in our applications.
Gingerbread usage is down from our last post to 11.5%; this is slightly lower than Google’s global number. However, KitKat usage is now at 16.3%, which is double Google’s global number. Overall, devices running Android 4.0 and later (for you minSdkVersion=”14″ folks) is at 87% of users; which is slightly above Google’s global count of 82.7%.
From my perspective, every time I write one of these, I’m left longing for more control over the metrics provided by Google Play. Something as simple as allowing us to read the cross of two metrics would be infinitely helpful in making more informed decisions about platform support. So 11% of my users are still running Android 2.3.3. Are they the same 11% that are running an outdated version of my application? If they haven’t updated my app in six months, are they worth my continued support? Are they even still using the application? Some of these can only be answered with targeted analytics, but many are questions Google could easily address with some simple tweaks to the metrics and dimensions filtering that would make a huge impact.
We are now six months past the announcement of Android 4.4 (KitKat) and quickly approaching the second birthday of Android 4.1 (Jelly Bean). We are seeing a higher rate of user adoption to the latest Android version after six months than ever before, which should be encouraging to all developers. But, at least with the data we have, it seems Gingerbread is not disappearing as quickly as many of us (myself included) expected. At Double Encore, our default position is to begin support at API Level 15 (Android 4.0.3) for current projects. In reasonably justifiable cases, we will address the effort vs. benefit of regressing that support back to an earlier platform. “Look at the size of the pie piece” is no longer reason enough for us.
Of course, we would love to hear from you as well. If you’ve posted your own application’s data, let us know in the comments or send us a link via Twitter, Facebook, Google+! Carrier Pigeon is also still accepted.
Want more insights like this? Subscribe to our email list at the bottom of the page (below the comments)!