The analytics example at least is really flakey ... plenty of web analytics companies didn't close up shop after Google's transformation of Urchin, a few years later they're still running and new ones are still emerging.
Google fails at a lot more than they succeed at. Chances are very good if Google enters a market they'll suck at it, and if you excel at it maybe they'll buy you if they want to own that market.
All you even have to do to beat them is provide any way for people to reach you for support.
Part of this is also due to the fact that the analytics business is so multi-faceted and had many valuable niches. That fact, combined with the increasing volume and variety of data people need to understand, makes the industry even more viable as time goes by.
Absolutely, I'm doing it for games and with quite some success.
But I think there's more to it than just that analytics is a particularly versatile industry, most of the stuff Google does could be retooled to solve x (or a variation of x) for a different y - and has: redtube, 5mins etc - let alone the also viable same y but better solution.
If it's not one of their handful of monolithic and to date unbeatable successes then their presence in a market really doesn't do anything more than validate it.
Google fails at a lot more than they succeed at. Chances are very good if Google enters a market they'll suck at it, and if you excel at it maybe they'll buy you if they want to own that market.
All you even have to do to beat them is provide any way for people to reach you for support.