ReadWriteWeb has picked up the story on Facebook application fatigue that I wrote on the weekend. The culprit I pointed at was the use of forced invitations. Some applications demand that users invite others at install time, before being allowed to use the application.
That blog piece generated quite a bit of debate on the Facebook developers forum. Many of the debaters admit that forced invitations are distasteful to them personally. A smaller group defend the use of forced invitations, because they appear to be a successful short term tactic for creating momentum around your application. Others argued that there is no proof that forced invitations are harming the Facebook ecosystem, and that there is no proof that people are using fewer Facebook applications.
To prove or disprove that claim, I ran a series of polls using Facebook's polling application. Facebook application developer Chris Diraddo also ran a poll. Here are the results:
The first think you will note is wide variation in the results. Because of limitations in the sample size allowed by the polling application, these numbers are accurate only to +/- 9.8%, 95% of the time. That is to say, there's a variation of 10% on average, and 1 in 20 times, a complete anomalous result will be produced. In addition, emotive bias may color these results. Emotive bias is caused by only users who passionately care about an issue responding to a poll.
Nevertheless, I find the results compelling. Polls 1 and 3 show a 56% and 61% refusal to install applications that require users to invite others. Both of these offer the user the option to give a nuanced "it depends" answer, as well. Many users may opt to use the app even if it requires invites, because of what the app does. For these folks, forced invites aren't necessarily a black and white issue. Poll 2, which forces a simple yes/no response, frames the issue in black and white terms. One interpretation of these apparently contradictory results might be that while many people dislike forced invites enough to state categorically that they wouldn't use an application which required them, in reality their behaviour is different.
Nevertheless, we can conclude that the majority of Facebook users (even taking into account the woefully wide confidence interval of the poll) would refuse to install an application that required an invite.
The most contentious argument is that the actions of the small (but growing) number of developers using this tactic has harmed developers by turning users of Facebook applications in general. Over the weekend, I published statistics from the Adonomics leaderboard showing that the top applications are seeing a pronounced decline in daily usage. In addition, numerous Facebook users freely admit that they now ignore "invite spam" from most applications. It seems that the population of Facebook users is now becoming fatigued with applications.
While it would be difficult for me to do an analysis across the entire universe of Facebook applications to see which, if any, applications are experiencing a downturn, I can use the polling technique once more to find out whether users are using more or less applications than previously.
In October, Compete.com published August Facebook data showing that out of 22 million visitors in the month, 14 million interacted with applications. That's nearly 64% of users. It also appeared that throughout the early part of the fall, application growth continued.
Fast forward to yesterday, when I asked a very simple question: Do you install Facebook applications? These were the results.
Again, because of the regrettably small sample size that Facebook polls restrict users to, this data is only accurate to +/- 9.8%. As many as 60% of Facebook users may install applications, or as few as 40%. Even so, there has been a drop in application installation since August.
Could that be solely as a result of forced invites? Not likely. After all, growing awareness of privacy issues in some applications is likely also a contributor. However, any tactic as widely vilified as forced invitations is likely to turn off the user base as a whole, and it seems clear that we're seeing a downturn in application usage. Perhaps it's not an absolute downturn. Many applications are still growing. Generally speaking, however, it seems that applications users are reaching the limits of their tolerance for fluff widgets and abusive marketing tactics.