What percentage of a Twitter feed can legitimately be fakes?
October 26, 2015 4:31 PM Subscribe
Popular Twitter feeds collect fake followers. What percentage would you think reasonable for bot followers, over which you'd suspect someone of having padded out their followers list?
I just ran my blog's feed through Twitter Audit and I have 5,694 real followers, 813 fakes, making 12% of my followers bogus. I've never attempted to inflate my list and nobody has said I did – it's just an example – but a recent report in my city is pointing the finger at various politicians on this issue, and I wondered, if you're in the hundreds of thousands of followers league, what percentage would start to look iffy.
I haven't found any articles mentioning this, but educated guesses would also be welcome.
I just ran my blog's feed through Twitter Audit and I have 5,694 real followers, 813 fakes, making 12% of my followers bogus. I've never attempted to inflate my list and nobody has said I did – it's just an example – but a recent report in my city is pointing the finger at various politicians on this issue, and I wondered, if you're in the hundreds of thousands of followers league, what percentage would start to look iffy.
I haven't found any articles mentioning this, but educated guesses would also be welcome.
Just ran that test on my account. 18K followers, never tried to inflate, and 13% "fakes."
posted by Conrad Cornelius o'Donald o'Dell at 5:54 PM on October 26, 2015
posted by Conrad Cornelius o'Donald o'Dell at 5:54 PM on October 26, 2015
The reporter who wrote that article about the politicians is being silly. People gather fake followers all the time through no fault of their own. I registered an account and I had like 12 followers within a couple hours, all "fake" accounts -- that is, robotic, automatic followers. The more followers an account has, the more fake followers they tend to get. It doesn't necessarily mean fake followers were bought.
There's also the non-fake, but equally insincere followers someone can gain by serial following. Basically, an account auto follows people in hopes that they follow back, and then the account unfollows them. I sometimes check out to see who unfollowed me and sometimes it people that must genuinely have gotten tired of my shit, but a lot of times it's serial following accounts that only followed me for a few days.
Funnily enough, a political organization I worked for was accused by some hostile bloggers of having fake Facebook followers. They said they checked our page's demographics, which is public, and the No. 1 city for our followers was Bangladesh and they accused us of buying fake followers. We never bought any fake followers, so we contacted Facebook and went through our follower history. There was a spike and, surprise surprise, the bloggers were pretty quick to notice. We think the bloggers bought the fake followers just so they could accuse us of "astroturfing."
At what percentage would I think someone bought followers? 50% or more, I suppose. But then again, I've never checked or would care to. If I look at sometimes timeline and they get few retweets and favorites, I assume they have a lot of followers that aren't actually interested in what they have to say. I get a lot of engagement on my tweets as long as I stay within a specific issue area, so I know I have legit followers who care about that issue.
posted by AppleTurnover at 11:54 PM on October 26, 2015
There's also the non-fake, but equally insincere followers someone can gain by serial following. Basically, an account auto follows people in hopes that they follow back, and then the account unfollows them. I sometimes check out to see who unfollowed me and sometimes it people that must genuinely have gotten tired of my shit, but a lot of times it's serial following accounts that only followed me for a few days.
Funnily enough, a political organization I worked for was accused by some hostile bloggers of having fake Facebook followers. They said they checked our page's demographics, which is public, and the No. 1 city for our followers was Bangladesh and they accused us of buying fake followers. We never bought any fake followers, so we contacted Facebook and went through our follower history. There was a spike and, surprise surprise, the bloggers were pretty quick to notice. We think the bloggers bought the fake followers just so they could accuse us of "astroturfing."
At what percentage would I think someone bought followers? 50% or more, I suppose. But then again, I've never checked or would care to. If I look at sometimes timeline and they get few retweets and favorites, I assume they have a lot of followers that aren't actually interested in what they have to say. I get a lot of engagement on my tweets as long as I stay within a specific issue area, so I know I have legit followers who care about that issue.
posted by AppleTurnover at 11:54 PM on October 26, 2015
It's not possible to answer this question with percentages, as fake followers are bought in groups of 1,000.
If you have 100 followers and buy 1,000 followers, you'll get a very different ratio than if you have 3,000 followers and buy another 3,000 followers.
posted by DarlingBri at 5:42 AM on October 27, 2015
If you have 100 followers and buy 1,000 followers, you'll get a very different ratio than if you have 3,000 followers and buy another 3,000 followers.
posted by DarlingBri at 5:42 AM on October 27, 2015
I worked for an organisation for blind and partially sighted people. I once ran my twitter through one of those checking sites and a lot of my followers that it flagged as fake were blind people. I guess because they were often still eggs, and didn't tweet photos. Even though they had lots of real content and conversations etc.... Whatever metrics they used, actual people slipped through, so I don't really trust them.
Also, if you do get followed by an egg, do double check its not a real person without a profile picture for some reason!
posted by Helga-woo at 7:37 AM on October 27, 2015 [2 favorites]
Also, if you do get followed by an egg, do double check its not a real person without a profile picture for some reason!
posted by Helga-woo at 7:37 AM on October 27, 2015 [2 favorites]
Most of my fake followers have icons of attractive women. I haven't done anything to court fake followers, and if I'm feeling feisty, I block/report them, but in the end I'm with AppleTurnover, it's a meaningless stat.
posted by advicepig at 7:58 AM on October 27, 2015
posted by advicepig at 7:58 AM on October 27, 2015
This thread is closed to new comments.
posted by xingcat at 4:48 PM on October 26, 2015