I always like to distinguish "with hindsight" from "at the time." Anyone can figure out what's strategic from hindsight.
At the time? I think probably Larry, Sergey, and Eric thought it was strategic then. Did they think it was more strategic than radio ads (a business long gone!) or TV ads (also gone!)? Not sure.
It was strategic. They needed to control the endpoints to protect search. Hence Chrome, Android, Daydream, watches and so on. Half of what Google does is an effort to control endpoints to protect search.
What I meant with "at the time" is: if you'd asked them THEN to distinguish a very few "strategic" projects from the lot of "not strategic," would they have picked Android?
Not "with benefit of analyzing all their moves since 2004, does a clear strategy emerge?"
The rule would have been "you can't say everything is strategic."
I actually think they would have refused to play that game. Because calling things "strategic" is stupid.
If you define strategy as choices/decision/capabilities which involve competitors, as opposed to "important", it doesn't have to be "very few projects". Very few strategic choices, but often a lot of projects to implement those choices. "Important" and "strategic" really aren't synonyms.
Walmart likely has 100's of projects under way which are focused on keeping prices low. They are all strategic because Walmart's strategic choice is to offer every day low prices. The probably have 10 projects underway related to updating their HR infrastructure - not strategic. Walmart's strategic choices are constrained - but once they made the choice to compete on price, hundreds of projects are strategic to them.
At Google, there has always been a fear/focus on the endpoints/clients. In Porter's terms, it's a fear of customer control, where the customer is a distributor rather than the end customer (Google search has about 3 or 4 customer sets, so it's not a perfect analogy). And it's a rational fear because if they don't control the browser or client, they are at risk of bing/others displacing them, or having to pay a fortune to be the default search engine in a browser (like they pay apple billions every year). There have been dozens of projects at Google which have more or less been "well if this is the interface of the future, we better be in it so we can keep search there". All these efforts are strategic. Sundar Pichai has been behind several of these efforts, because he gets it.
Calling Chrome "strategic" didn't sound stupid to anyone. Calling the self driving car "strategic" would have sounded stupid to everyone.
