As social media platforms tout the number of users as a value of their overall worth, the fact many ‘followers’ and ‘profiles’ are fake is illustrating a growing issue with spam and advertising analytics. In July the BBC set up a fake business ‘Virtual Bagel’, began advertising on Facebook, and determined that a large majority of likes came from South East Asia and were most likely fake accounts or bots. Another company, Limited Press, ran analytics on their own Facebook advertising alleging more than 80% of likes were fake.
Facebook itself was forced to reveal in company filings that 8.7% of over 955 million profiles broke rules in some way – duplicate profiles, misclassified accounts and undesirables – those who use accounts solely for spamming.
Newsday reported an analysis on Presidential Candidate Mitt Romney gaining more than 100 000 Twitter followers in one weekend, most of whom were fake. Twitter has yet to publish numbers on percentage of overall fake vs real users, but Tweet spam is a continual and growing headache for many users.
As social media platforms rely almost solely on advertising for revenue, their value to advertisers diminishes dramatically when companies discover ‘likes’ and clicks on paid links are worthless.
A new service, Status People, will parse your personal and business Twitter accounts for fake users, and give you a percentage report. While it is not an exact science the algorithm has some guiding principles, and becomes more accurate as more users check their followers:
“Spam accounts tend to have few or no followers and few or no tweets,” the company writes. “But they tend to follow a lot of other accounts.”
You can then use a system like TwitCleaner to identify fake, bot or inactive users that you either follow, or follow you, and clean up your account.