Can Microsoft switch from Windows Phone to Android?

Can Microsoft make an Android device? This question has been bothering the press ever since Microsoft announced it's getting a new chief executive, Satya Nadella, and with leaks of an entry-level Nokia smartphone running on Android, the improbable has suddenly become the hot topic.

Taking existing Android and building on top of it, or ‘forking’ it, is exactly the option that Microsoft has when it comes to Android. There are plenty of reasons in favor of a Microsoft fork of Android, a Microsoft/Nokia Kindle of sorts, and probably as many or more against, but we can sum up the argument down to this: Windows Phone has not been capable of attracting sufficient sales to break the Android/iOS monopoly and create a true ‘three-horse’ race, so that's why an Android smartphone by Nokia (or Microsoft) would leverage a much larger app and developer ecosystem. Finally, since parts of Android are open source, it's not risking too much.

On the flip side of things, it would be an admission of failure of the Windows Phone platform, something that we doubt Microsoft would be willing to do despite the sales situation. Also, it would require some effort and a cloud back-end infrastructure. Let’s be theoretical for a minute and try to step in both Microsoft’s and Google’s shoes to see what it would take for a Microsoft ‘Droid’ to become a reality.

The one argument in favor of a Microsoft ‘Droid’: market share

2013 is long over, and with its passing, we can safely say that the mobile space continues to be a duopoly of two operating systems - Android and iOS. Microsoft’s Windows Phone has a very marginal market share of around 3.2%, according to data for 2013 from market researchers Canalys); in the United States, Windows Phone share is even lower, at 3.1%, and some analysts like the Microsoft-focused Paul Thurott and the AdDuplex network estimate the total platform user base at around 50 million. In stark contrast, Android has an estimated total installed user base of some 1.9 billion, while iOS and Mac combined have around 680 million, so while Microsoft is indeed the third-largest platform, it is so far behind Android and iOS that we cannot seriously consider it a ‘three-horse’ race just yet.

Even worse, the latest statistics show that Windows Phone grows to be more and more dependent on a single phone maker - Nokia. Windows Phone devices by the Finnish company account for more than 90% of all Windows Phone sales. We are all hoping that Microsoft will announce new partners soon, but the reality of the situation at the moment is that its platform has not been met with enthusiasm by phone makers other than Nokia, and the huge sales are not there to give more people reason to buy in.

Additionally, statistics also show that the dynamics of Windows Phone are actually moving down, with diminishing interest in Windows Phone smartphones, and low sales for the platform’s top-shelf smartphones like the Nokia Lumia 1020. What happens instead is that low-end devices like the Lumia 520 that bring low profitability for phone makers are growing their relevant share of the Windows Phone market. That cannot be a good thing for the manufacturer’s revenue sheets.

All in all, the first 'app argument' in favor of an Android-based Microsoft/Nokia could be the strongest, as the Windows Phone ecosystem seems to have been ‘too little, too late’.

Why Android: because of open source, or because of developers, developers, developers?

If Microsoft were to call it quits on Windows Phone, out of all mobile operating systems out there, Android is the only open source one that is a viable replacement option. Smaller ones like Ubuntu have even less market share than Windows Phone and are clearly not a good fit, and iOS is a proprietary platform, of course.

With over 1 billion applications in its catalog, and a strong developer ecosystem, Android can provide Microsoft with a base that it’s currently missing with Windows Phone. Moreover, Android has largely maintained the principle of providing a level playing field for third-party manufacturers, so that it has the APIs and all other needed tools for Microsoft services. And a quick look at Microsoft’s revenue sheet will tell you in perfect clarity that services like Office are the exact reason why Microsoft wants to be on mobile.

It’s important to note that Android’s openness allowed even starting this discussion about a Microsoft ‘Droid’. Google started with the Android Open Source Project (AOSP)’s huge codebase launched in the early days of Android around October 2008, and has been updating it since then. AOSP is basically the bread and butter of Android, something that a third-party phone maker can virtually take and slap on a device, and have a fully functioning smartphone. It has literally all that’s needed - the launcher, the dialer, phone and contacts apps, the calendar, camera, gallery and so on.

It’s also equally important to make it clear that while the AOSP provides the backbone of Android, it does not include things like Google’s excellent suite of applications including the Google Play Store, Maps, Drive, Gmail, Chrome and so on. All of that is packaged in a separate entity that’s often called Google Mobile Service (GMS), or Google Play services. GMS - unlike AOSP - is proprietary, and has been so from the start, designed to provide access to Google’s cloud services on top of Android. Making use of it is something that requires a paid validation that is rumored to vary in cost, but average at around $0.75 per device.

One key feature of the GMS is that it cannot be divided, you cannot have just a part of it - it is a buy-all, get-all solution. In practical terms, this translates into the fact that all GMS Android phones have not just say Google Play on them, but also the full suite of Google apps.

Here is where the big question for Microsoft appears. If - purely theoretically - the company was to build an Android fork, would it want one that is based on AOSP only or one with AOSP and with GMS?

Like Amazon...

The first option is actually something that has already been tried and proved to be possible with Amazon’s Kindle Fire tablets, built under this model. For Microsoft to build a platform on top of AOSP only would mean no access to the Google Play Store and its huge catalog, so it’d have to build an app store from scratch like Amazon did. It might make sense to some since such a store would be based on Android, and developers shouldn’t have much of a trouble bringing existing Android apps to it, but it is such a grand undertaking that to us it simply does not seem viable in this day and age of already mature app catalogs.

Or like Samsung?