Social-networking site Facebook saw an 89% increase in U.S.-based visitors in May compared with the same period last year, according to Internet metrics firm ComScore.
The site saw 26.6 million unique visitors in the United States, a result that suggests Facebook's decision to drop registration restrictions in September was the right one.
Before it was open to the general public, Facebook had about 14 million unique monthly visitors.
In late May, Facebook opened itself up as a platform for third-party developers, offering them limited access to its data and infrastructure to build and market their own applications to site users.
At the San Francisco developer event where that announcement was made, CEO Mark Zuckerberg pitched the Facebook platform as a place to profit. "You can build a real advertising business off of Facebook," he told developers. "If you don't want to run ads, you can go ahead and sell something."
Comparing May 2006 to May 2007, Facebook also saw its average number of monthly page views jump from 6.5 billion to 15.8 billion, a 143% increase. The average number of minutes spent on the site per visitor per month also increased, rising from 138 to 186 on average, a 35% increase.
During the same period, MySpace also saw its audience grow, though not as quickly. It went from 51 million unique visitors in the United States to 69 million, according to comScore, a 34% increase.
The average number of monthly page views on MySpace went from 29 billion to 44 billion, a 51% increase. And the average number of minutes spent by site visitors rose from 186 to 196, a 5% increase.
As a consequence of opening up, Facebook has been attracting visitors from outside its 18- to 24-year-old user base. Comparing May last year and this year, Facebook saw the most growth among 25- to 34-year-olds (181%), those between 12 and 17 (149%), and those over 35 (98%). The site's traditional 18- to 24-year-old demographic grew by 38%.
Another ratings company, Nielsen//NetRatings last week said that over the past six months, Facebook's audience in the United Kingdom grew at 19 times the rate of growth seen by MySpace, 523% compared to 28%.
But such numbers should be viewed with some skepticism. ComScore itself has acknowledged that three in 10 computer users regularly delete cookies -- files used for visitor tracking -- from their Web browsers, resulting in a potential 150% overstatement of audience size by those tracking such data. (ComScore published a study about this to support its case for using alternative counting methods such as audience panels.)
In April, the Internet Advertising Bureau published an open letter taking ComScore and Nielsen//NetRatings to task for inaccurate, antiquated methods of measuring Internet traffic, such as audience panels. While the two companies and the IAB have since patched things up, accurate Internet audience measurement remains something of a dark art.
And then there's the larger issue of whether the companies gathering such data can be trusted. A recent report by spyware researcher Ben Edelman says that ComScore doesn't always get consent to install its tracking software. "ComScore pays independent distributors to install comScore software onto users' computers," Edelman explained in his report. "Predictably, some of these distributors install ComScore software without getting user consent."
Granted, the willingness to profit from spyware doesn't necessarily indicate bad math, but it does raise some questions.