I don't know if there is an example of a public internet company that was able to transition from being an ad-base business into selling stuff. The closest may be LinkedIn, who's selling recruiter tools and premium memberships. However, LinkedIn had been doing that before the IPO, as a relatively small company.
I'm skeptical that Facebook can pull this off in a reasonable time frame, but shareholders who are psychologically anchored to the IPO price are probably more optimistic.
Which sounds remarkably similar to the way people thought about companies during the dotcom boom: Get the users first, figure out how to make money off of them later. Facebook's admittedly got one key difference from companies during the dotcom heyday, which is that instead of a burn rate it has this thing called 'revenue'. But that aside, at least for a while investors seemed to have fallen back into the old trap of thinking that the monetary value of a user isn't somehow tied to the amount of income they provide.
Maybe Facebook will figure out something else. . . but if nobody's sure what that is right now then it's still an Underpants Gnomes[1] business model.
I agree with your general premise however, that Facebook's primary growth won't be on the user base, but needs to be on the revenue per subscriber numbers. This is going to really test one of the Valley's most recent premises, that building a massive user base will also create a massive business (Twitter, Instagram I'm looking at you...)
[1] http://www.telegraph.co.uk/technology/facebook/8573340/Faceb...
[2] http://www.forbes.com/sites/ericsavitz/2012/05/09/facebook-t...
That's a bit over 10% of the world's population. They can't really get the growth that's expected with their P/E, especially because of the difficulties that they're having with getting higher returns/user. They can't grow their userbase by that much either.