You don't want speculation, rumours and tales. You want answers. Which is good, because that's exactly what you're going to get in this, the pro guide to working out who's been buying Twitter followers. This, is the stuff that some of Twitter’s most mouthy, boastful and smug users do not want you to know…
WHAT IS BUYING FOLLOWERS?
“Buying followers”, as defined for the purpose of this post, is any means by which a Twitter user has paid actual money to gain follow button clicks. This could be anything from a lump sum exchanged for a large quantity of mechanical bot follows, to a directory subscription which artifically incentivises real people to follow an account they wouldn’t otherwise care about.
As I explained in my Why Twiends Doesn’t Work article, the effect is pretty much the same whether the followers are real people or mechanical bots: big badge of honour, no one actually listening.
So let’s get started with our ten signs of artificially purchased Twitter followers. It’s best to think of this list as a compound indictment, and cross-reference the various points against each other. The more of these boxes you can tick, the more likely it is that the user has been buying followers…
1. HIGHLY IRREGULAR INCREASE(S) IN FOLLOWERS TOTAL
This was always going to be the most obvious starting point. If you notice someone’s followers total increasing at a steady crawl, then skyrocketing by many thousands in the space of a week or two, and then returning to the steady crawl again, you can bet pretty safely that they’ve just bought a load of followers.
It would take a spectacular and sustained burst of 'real-world' publicity to effect this sort of rocket-boost without artificial inducement or mechanical following. So unless you can establish that the user has had sustained, high-profile TV/media coverage, jumps like this will almost always indicate fake follows.
2. RETROSPECTIVELY DETECTING IRREGULARITIES IN FOLLOWS
But what if you don't notice the jump when it happens? Can you still detect irregularities in someone’s follow rate at a later date? Yes you can. Assuming the account is active and tweeting, watch the followers total for one week (you can do it over 24 hours, but it’s not as reliable). Note how many new followers the account gains.
Next, note the user’s join date. If there’s no join date visible, see my How to Find a Twitter User’s Join Date When It’s Not Displayed post on Twirpz.
Now simply work out how many weeks the account has been open, and divide the user’s total number of followers by that number of weeks – giving yourself a weekly average.
How does the weekly average figure compare with the number of followers you’ve seen the user gain in the last week? If the average figure is massively higher, the user’s followers total has probably seen a big jump at some time, and thus, they may well have bought followers.
- Bob has 100,000 followers.
- His account was opened on 3rd January 2015 (today is 3rd January 2016).
- He gained 24 followers in the past week.
So… 100,000 (Bob’s total of followers) divided by 52 (the number of weeks Bob’s account has been open) = 1923.
Bob’s weekly average gain in followers is 1923, but last week he only gained 24? Doesn’t add up, does it? The likelihood is that Bob’s total increased with at least one irregular jump in the past. Hence, there's a strong chance that Bob has bought followers.
3. DISPROPORTIONATELY LOW ENGAGEMENT
When someone’s Twitter followers are real, some of them will engage with the user. They’ll Like and perhaps Retweet, and most importantly, they’ll Reply. But when someone buys Twitter followers, there’ll be a vast, disproportionate lack of response to their tweets.
You should track engagement as a whole, rather than just looking at one aspect such as Likes or Retweets. Individual aspects can be fudged. People can, for example, Like their own tweets using alternative accounts, etc. But the combined picture of engagement – the Likes, RTs, @replies and @mentions is extremely difficult to fake.
For a user with a huge followers total, first check that they're tweeting regularly. Provided they are, type the command below into the Twitter Seach box…
Replace @PersonsUsername with the actual Twitter username of the individual you’re checking out. Don’t change anything else. Don’t add any spaces. If you're using the desktop version of Twitter, click the Tweets selection in the More options dropdown.
This shows you the volume of @replies the account holder is getting. Accounts whose followers are entirely, or almost entirely bought, have incredibly low @reply volumes – often as low as one or two replies per month. If you do see @replies, check whether they’re all coming from different people, or just coming from the same one or two accounts. Accounts with very large numbers of real followers will get a stream of @replies from a wide range of different accounts. Accounts whose followers are fake will either get very few @replies, or all the @replies coming from the same account(s) – their own alternative accounts, essentially.
Don’t forget to reality check the totals of Likes and RTs against the totals of @replies. For example, is it realistic for a tweet to get 200 Likes, but no @replies?
Ultimately, if you have 50,000 followers, you issue 10 tweets per day, and you get one @reply per week, your followers are extremely unlikely to be real.
StatusPeople will automatically furnish you with a report for a given Twitter user, estimating how many of their followers are fake. This is a useful resource, but as with any other measure, it should not be considered a standalone solution.
The accuracy of StatusPeople falls down when a lot of followers are readers as opposed to talkers. The site states…
“On a very basic level spam accounts tend to have few or no followers and few or no tweets. But in contrast they tend to follow a lot of other accounts”.
However, if you look at big, successful entertainment accounts, many of their followers conform to these criteria. You see lots of ‘egg’ profiles who don’t want to talk or be identified – they just want to be entertained. And because big Twitter presences with a ‘zeitgeist factor’ tend not to follow back, people who ONLY follow that type of account, or just follow celebrities, etc, will have very few followers. So even though they don’t talk or identify themselves, these followers can be very valuable. If you’re marketing something, it doesn't really matter if not everyone responds. It's most important that people listen, and a lot of these accounts do listen. StatusPeople kind of clauses in this factor by alluding separately to accounts with over 50,000 followers (in other words, celebs and the like). But 'readers/listeners' will follow smaller accounts if they're giving good value, and some minor celebs, who often get a lot of these 'listen-only' accounts following them, may only have 1,000 to 5,000 followers.
I tested StatusPeople with the Twitter account of someone who is in the public eye, and whom I know has not bought followers. In the test, the user’s follower stats came up as 35% fake, 39% inactive, and just 26% good. Knowing that this person uses Twitter to drive traffic to another site, and having seen verification of the very large number of hits they get on that other site as a consequence, I would very seriously doubt that StatusPeople’s prediction was even vaguely accurate in this instance. The “35% fake” is very misleading, because this is someone with a high public profile who blatantly has not at any stage seen a need to artificially inflate their follower stats, and actually only has quite a modest followers count compared to other users with the same source of 'real world' publicity. There are undeniably loads of silent 'eggs' there, but they're real people. Their follow lists are very uniform in terms of genre and interests, and you just don't get that with bought followers. Bought followers follow whoever pays, so their follow lists are much more random.
Conversely, there are users I know have bought followers, but who score well on StatusPeople. The results will often depend on where you ‘bought’ your followers. For instance, pay Twiends or another directory-type resource to induce real people to follow you, and StatusPeople will score you well. But those artificially-induced followers will likely not pay any more attention to you than a collection of bots. Some vendors of mechanical follows are getting more sophisticated today too, and can fudge interactivity to the satisfaction of sites like StatusPeople. They can get their bots to follow each other, auto-tweet, auto-reply and Like each other’s tweets – thus, again, obtaining good scores when in reality the followers are outright fake.
But with all that said, there are instances where StatusPeople can be extremely accurate – especially when the proportion of fake followers is dramatically dominant. If the report shows an overwhelming number of fake followers (like 70% plus), I’d be inclined to regard that as pretty conclusive evidence that the user has bought at least some of them.
5. DISPROPORTIONATE NUMBERS OF FOREIGN FOLLOWERS
Some of the entities supplying bought followers are not based in the same country as the user making the purchase. So often, the blocks of fake followers, as they arrive, instantly look out of place – particularly if their accounts are written in a different language. Look through the user’s followers for blocks of foreign usernames, and evidence of foreign language in the text.
Tip: You can use a resource like Tweepi to dive in at set points in an account’s follower history, rather than just having to scroll through all followers. A lot of users buy followers when the account is new, but then bury those fake followers under their subsequent real followers as time goes on. Using Tweepi you can quickly call up the first or early blocks of followers for a given account.
6. NO BIO (OR SOME INSTANCES OF REPEAT BIO TEXT)
Whilst you’re looking through your suspect’s followers, check for blocks of accounts with no bio entered. Not all real users enter a bio, obviously, but a long block of followers without any bio (particularly if the accounts have similar characteristics) is a classic indication that someone’s been buying them in. The people selling these mechanical, artificial followers have to create thousands of accounts, and it’s impractical for them to spend time considering the bio text. Sometimes, if they bother enterting bio text at all, it’ll be cut and paste lines that get repeated across many accounts – so look out for bio text repetitions too.
7. BLOCKS OF FOLLOWERS WHO HAVE STOLEN PROFILE PICS
Most of the ‘businesses’ creating fake followers to sell on, will have to steal the profile pics for their accounts. Clearly, they’re not going to commission a photographer to take high quality portraits for them, so if they’re using high quality portraits, they’re going to be nicked. If you come across a block of followers that look suspicious (gorgeous profile pic portrait, but zero bio and just 5 followers, for example), you can check out the profile pics using Google or Tineye’s reverse image search capabilities (see How to Prove a Profile Pic is Fake). These routines should give you a good idea who the real owner of the profile picture is. Reverse image search is also a very quick process once you know how to do it. I normally check three profile pics from a suspicious block – if they’re all someone else’s pictures, the block of followers is probably bought.
Has the user ever been accused of buying followers? Enter this text into the Twitter search box…
@PersonsUsername bought followers
@PersonsUsername buy followers
This will sometimes reveal accusations made by other Twitter users. Obviously, an accusation in itself does not confirm that someone has bought Twitter followers. The accusation may, however, be accompanied by some reasoning and some proof. You should also note if, and how, the accused responds. Those who haven’t bought followers will typically respond with a forthright and convincing denial. Those who have, more usually tend to ignore or block the accusor.
You can widen this search further by simply entering...
This can sometimes reveal unexpected info, and it can exonerate an account holder as well as implicate them. Just see what comes up in the search. The person might have been very proactive in sourcing followers, which would mean they probably didn't buy. Or alternatively they might actually quote their followers total in tweets from time to time. Or perhaps a (real) follower will congratulate the user on reaching a followers total. I used this method to check out one user, whose total turned out to have increased by about 150,000 within a few weeks. It’s two and a half years since that happened, but despite full, continued activity, the total has only grown by another 10,000 or so over that much longer period. Method number 2 would not have presented such clear-cut evidence of an irregular jump in this particular case.
9. LOW TWEET COUNT, LOW FRIEND COUNT, HIGH FOLLOWER COUNT
If a user has posted, say, just 15 tweets, and is following only 50 friends, but has 40,000 followers, then they either have a huge existing public profile away from Twitter, or there’s something fishy going on. It’s pretty well understood that a tweet total in itself doesn’t equate to followers – users can employ so-called “churn” tactics (strategically mass following and unfollowing) to inflate their stats without tweeting at all. So this doesn’t in itself mean the followers are bought. But it’s certainly an alarm bell, and should be considered alongside other factors.
If you’ve never heard of the account holder, Google their name. Are they famous? If it appears not, look for a website link in their Twitter bio. If there’s a link there, DON’T click it! Copy it and paste it into a search engine, and see what comes up. Match the information you get with the info you find within the Twitter account, always remaining sceptical of catfish traits.
If you haven’t been literally hammered with information about this person after those two web searches, you can conclude they don’t have a public profile elsewhere, and whether bought or not, their followers cannot be considered ‘real’.
Remember though, this would not typically apply if the account holder is following a large number of ‘friends’. If someone has 40,000 followers and is following 35,000 accounts back, the likelihood is that they just proactively follow for reciprocation. I wouldn’t associate an account that has vaguely equal follower/friend totals with buying followers - even if they hardly ever tweet.
FINAL REALITY CHECK
This is where you stop and say: “Seriously… does this total of followers actually make sense, for this person?" Your local grocer has 200K followers? Of course he doesn't. If his account is showing 200K followers, he’s not going to pass a reality check, because why on earth would 200,000 people follow your local grocer on Twitter? Unless he’s made a phenomenally viral video or similar, they’ve no idea who he is. And even if he has made a phenomenally viral video, the likelihood of that translating into 200K Twitter follows would be extremely remote.
As part of your final reality check, consider…
- How likely would YOU be to follow the account? Is it good value or just rubbish and spam?
- How did people find out about the account? Where did the publicity come from? And if you can answer that, do other users with the same source of publicity have a followers total in a similar ballpark?
- How can someone who is not capable of attracting any audience outside of Twitter, manage to attract such a vast fanbase on Twitter?
People who buy Twitter followers are really only kidding themselves. The problem is, some use their perceived 'superiority' to control or belittle other users - perhaps even bully them. In the end, if someone just feels better about themselves because they have a "350K" denotation beneath the word "Followers" on their Twitter profile, that's fine. But if they try to exploit or subordinate the public with it, that's a problem. The practice of buying Twitter followers has shown up in all walks of life, from the local smart-arse, up through big business and massive celebrity, to government. You don't have to be a control freak to buy Twitter followers, but in my opinion, the probability is that you will be. Don't get too cocky, though. Some of us know what you've been up to.