It’s been about 10 days since Facebook debuted its applications and the clear winner is iLike, which has well over a million users. It’s taking turns with the Mobile platform at the top of the most-users list:

Last.fm, who lost a week while getting their app ready to go, are rising quickly but far far behind with just over 40,000 users:

If you’re REALLY bored it’s kind of fun to spend a couple of minutes looking at the ‘most used applications’ list. You can refresh continuously and watch the numbers climb. Last night Last.fm was picking up about 5 users per refresh while iLike was picking up around 20.

MOG’s application which launched the same day as iLike’s, meanwhile, is floundering. It peaked at just over 20,000 users and now seems to be losing users at a rapid rate. This morning Facebook says it has about 5,400 users. Ouch Update: In comments David Jennings points out that there are 2 MOG applications, and the one that had 20,000 users now has around 22,000. My error. Sorry and thanks for the catch David!

I wondered what implications the success of a Facebook application has on the website itself. So I bopped over to Alexa. I don’t trust their numbers very much in an absolute sense, but as a means of comparing one site to another at least most of the biases are going to be comparable (though I wonder about the international distribution of Alexa’s data points and how that affects this). Anyway, it’s the best data we’ve got I think, flaws and all.

When you compare these three sites over the last three months, it looks suspiciously as though all those new users on Facebook might be pulling users away from iLike.com itself. Below are the comparative graphs for iLike, Last.fm, and MOG in terms of reach and rank. You can see that Last.fm is higher than the others (though the difference is much less than it used to be), but look at what happened to iLike in recent days. Not only does it seem to be dropping dramatically, MOG has overtaken it in rank.

As a social scientist I know better than to conflate correlation with causality. There could be several interpretations of these figures. But it certainly makes me wonder if iLike users are switching to using the application within Facebook rather than going to the iLike site. Some sites, like ReverbNation, have designed themselves to be most successful when spread throughout the internet. Is iLike such a site? What are the implications for destination sites if integration with other destination sites means loss of their own traffic?

I think we will see more and more of this sort of merging of multiple sites into one at user discretion. Pageflakes, iGoogle, and other start pages are doing this. I waxed about my dream of doing this (only better) here. But it requires new ways of thinking about what it means to be a successful internet company.