Google has invested much in Android, dumping in huge volumes of engineering, design and marketing effort. They did this in reasonable expectation of future benefit, in the form largely of mobile advertising revenue and commission on app sales. They receive an ancillary benefit through ensuring that the web remains open on (at least some) mobile devices, making it harder for competitors’ platforms to freeze them out.
The primary weakness in this model is that people can take all the work Google’s done (here’s the recipe), replace Google’s apps and sell it.
Google’s prevention mechanism is to wield the Android trademark and the use of Google’s apps (mail, maps, etc) as a carrot, via the controversial compatibility test. This works because everyone involved believes that (a) consumers recognise the Android brand as a value symbol, and (b) the utility provided by Google’s local apps would be hard to replace.
Amazon’s Kindle Fire, will likely demonstrate for the first time that it’s possible for a competitor to take Android, strip out all Google’s carrots, and make real money in the mass market. Worse, they’ve established a competitor marketplace, allowing others to make use of the underlying platform but completely avoid Google. If this comes to pass, Google should be very worried; they’ve made the upfront capital investments to build this platform, but they would then be at real risk of never realising the long-run revenue on which their business case was presumably predicated. Their saving grace here is that Amazon probably doesn’t have good alternatives to Google’s applications available; they won’t need them because the Kindle Fire is a narrower device, not a general purpose tablet.
If an alternative supplier of the functionality in Google’s apps emerges (Microsoft and Yahoo!, I’m looking at you), either independently or in conjunction with Amazon, they could reasonably set up a competing ecosystem, but continue to benefit from Google’s engineering effort in the base platform. Worse, the more profitable Android becomes for Google, the more attractive this approach becomes for those who would cuckold it. Yahoo has nothing to lose; Microsoft’s only reason to stay away from this path is the potential for cannibalising Windows Phone’s revenues (although they already make money on some Android devices), unless they can successfully differentiate the two offerings.
Google’s remaining primary competitive advantage (vs. “Clonedroid”) in this universe would be their application teams’ early access to new Android functionality, but since most Androids never get updated, actually wielding this seems very difficult.
The remaining question, then, is what the carriers want. The carriers are the primary channel by which consumers get access to new devices, and they are generally assumed to keep tight control (hence the obvious discomfort at the level of control Apple has wrought from them). So long as Apple is ascendant, it’s in their interest to collaborate with Google to improve the alternatives; if Google ever becomes difficult for them to deal with, they have every reason to look elsewhere.
All of which means that Google’s position looks pretty weak, despite rocketing market share growth and current revenues.