#1 trumps everything else. Try doing the same in a market the size of a typical European nation and you end up with an embarrassing digital backwater. Do it in a market the size of China and you can trail silicon valley by a few years at most, constantly gaining relative competitiveness as the global market of "online stuff" slows from wild, exiting exploration to establishment.

I suspect that #2 and #3 are far behind what I would put on #2, plain old language barrier, but I am surely no China expert.