Twitter killed off spam accounts in run-up to IPO
In the months before its IPO filing, Twitter worked hard to purge fake and spam accounts in order to provide advertisers with realistic user numbers. But you can still buy fake followers.
In the months preceding its IPO filing, Twitter worked hard to purge fake and spam accounts, long a thorn in the company's side and an impediment to its ability to attract advertisers.
Given that it's so easy to register for Twitter, and that there are few restrictions on sign-ups, there has been little to stop spammers from creating large numbers of accounts. And the company has few illusions that the problem isn't significant.
In its IPO filing Thursday, Twitter identified spam as one of the significant risks that could affect the company's business. "Spam could diminish the user experience on our platform," the company wrote in its S-1, "which could damage our reputation and deter our current and potential users from using our products and services."
Over time, many users have complained that large numbers of their Twitter followers are fake. For example, Popular Mechanics writer Douglas Main concluded in a 2011 analysis that 19 percent of his Twitter followers were fake, or spam. Even worse, Main's wrote, his colleague John Herrman's roster of followers was plagued by 48.6 percent spam or fake accounts.
Of course, not everyone abhors the practice. Last year, a report surfaced that as many as 15 percent of Republican presidential nominee Mitt Romney's Twitter followers may have been fake. Although it wasn't clear who purchased the followers, which made it look like the candidate was more popular on Twitter than he actually was, the Romney campaign denied having been involved.
But Twitter knows that to attract advertisers, and to get the maximum advertising value it can, it has to clean up the service and convince marketers that they are reaching humans.
And that's why Twitter has been actively working to purge such accounts from the service. In its IPO filing, the company noted that, "We are continually seeking to improve our ability to estimate the total number of spam accounts and eliminate them from the calculation of our active users."
In addition, Twitter said that, "We made an improvement in our spam detection capabilities in the second quarter of 2013 and suspended a large number of accounts. [And] spam accounts that we have identified are not included in the active user numbers presented" in the S-1.
Though it did not clarify how it was identifying the illegitimate accounts, or how quickly it was removing them, there's no doubt that proactively slashing spam and fake users was an essential step for Twitter to take as it works to maximize its value in advance of its IPO. Though there's no way to know exactly how many accounts the company killed off, it said in the prospectus that there are now less than 5 percent of monthly active users it considered to be spam, or fake. That number appears to be lower than Facebook's total percentage of duplicate, misclassified, and undesirable accounts, which was reported to be 7.2 percent in that social network's most recent quarterly report.
But Twitter clearly hasn't fully solved the problem. Even today, in the wake of the purge, it's still easy to do precisely what Romney was accused of doing in 2012 -- buy followers. A quick Google search revealed that for $11, anyone can buy 1,000 followers.
