Back to blog

Android First is the New Normal

Mark Wilcox · August 05, 2016

The mobile platform landscape was fairly stable for more than two years. Having both won the platform wars, Android and iOS seemed quite settled into their market positions. Android selling the most units in every market, but with iOS taking a dominant share of the lucrative high-end. Similarly, Android’s greater developer mindshare was always counterbalanced by iOS developers making the most revenue, and iOS being the primary platform for more full-time professionals. In the last six months we’ve seen a very significant shift on that last point. Apple will now have to work extremely hard in the next few years to avoid giving up further ground.

android_first_illustration

Why developers are prioritising Android

Towards the end of 2013, Steve Cheney wrote a very widely-read post on Why Android First is a Myth. We wrote a response at the time highlighting the strong silicon valley bias that made the conclusions doubtful, but also confirming with our data that, at least, professional developers were quite heavily prioritising iOS. In 2014 we heard that Android first was a fallacy – the user base might be there but fragmentation, plus inferior documentation and tooling, would make it a poor trade-off for many. In 2015 debate on the topic continued while our data showed Android gradually winning the priority of professional developers from other platforms, but not iOS. We entered 2016 with Android marginally ahead: 40% of professional developers globally prioritised the platform versus 39% for iOS. That seems to have been a tipping point, with our latest survey showing a large shift from iOS to Android. A massive 47% of mobile developers now tell us they consider Android most important, while preference for iOS has slipped to 31%.

The reasons for this shift are many but related. The fastest growing regions in terms of mobile developers are those dominated by Android already. New mobile developers are increasingly choosing Android first in all regions. Existing developers are shifting their priority to Android because the types of app they build are changing. We’ve been highlighting for some time that the app economy is shifting away from direct monetisation of apps, to using apps as a channel for some other business. Mobile commerce is by far the largest and fastest growing (at least in terms of total revenues) part of the app economy. Demographics still matter – users of iOS spend more on real-world goods and services through each device than Android users. However, unlike with downloads and in-app purchases, the difference is nowhere near enough to make up for Android’s larger user base.

Leveling of the playing field

For startups trying to find product-market fit, and enterprises needing to deploy across multiple platforms, a key reason to build on iOS first has been the ability to get to market faster. Android had more relaxed deployment policies and much faster publishing cycles, but it was was easier to develop higher quality apps for iOS. The most critical area in this regard has been the UI. Historically iOS had a few fixed resolutions and higher quality APIs for UI development. Android had a vast array of screen resolutions and aspect ratios, and fairly basic abstractions for dealing with that complexity. However, with the launch of the iPhone 6 and 6 plus in 2014, along with split screen mode for iPads in 2015, Apple has forced iOS developers down a scalable UI path as well. Once the UI of an app has to adapt to multiple resolutions dynamically, the complexity isn’t significantly increased by having to work with many more resolutions (maybe in testing but not in development). So iOS development has become harder, whilst Android has had time to mature their UI APIs, providing support libraries to reduce fragmentation across operating system versions.

The other area where a major Apple advantage has been eroded is in the design of app interfaces. Both Apple and Google moved towards a flatter and more minimalist style, which shifts the emphasis towards animation as the way to both give an interface personality and help users to understand it. Apple’s approach has been to show rather than tell, and leave artists to create, while Google has provided extensive guidance to developers on how to implement their Material Design. The latter has been favoured by many developers that don’t want completely custom UIs deciding to design for Android and adapt that for iOS, where the reverse was previously the case.
Apple isn’t doomed just yet
It’s important to note that mindshare for iOS is still within the range it has occupied for the last few years at 52%. Developers are not abandoning the platform. Shifting developer priorities towards Android are quite a long-leading indicator for device sales, and Apple are already making big moves to counter this trend. Swift is a boost to iOS developer productivity. Cutting app review times increases iteration speed. Apple Pay on the web gives iOS a greater advantage in the booming mobile commerce market.

The danger for Apple is if Android first becomes firmly established as the new normal. If some of the best apps and services come out on Android first then some of the early adopters will start to migrate. Where the early adopter user base goes, other users and the cool startups will follow. Without an app ecosystem advantage, Apple would become almost entirely dependent on hardware differentiation to maintain a price premium. This battle is far from over but Apple will be working a lot harder to keep developers focused on iOS than they have in the past.

The tip of the iceberg

We’ve just taken a deep dive into one interesting trend in the mobile developer ecosystem, there’s more going on with the mobile browser and Windows 10 too. We also have the latest trends from desktop, cloud and IoT, as well as new insights into augmented and virtual reality, plus data science and machine learning. Our State of the Developer Nation Q3 2016 report is filled with interesting trends as seen by 16,500+ developers across 145 countries.

Get it here