Do you ever come across a business profile or page and think 'what the...? How did they get 3,000 followers?' As with most things in life, if something seems fishy, there's a good likelihood that it probably is, and with the fake social media profile industry worth hundreds of millions of dollars per annum, it's not hugely surprising to find out many individuals and brands have taken this route. Like, a heap of them have - just take a look at the results from the recent Instagram fake profile purge, where a whole range of celebrities took big hits in their follower counts.
And it makes sense, having more followers and likes can definitely improve your brand position - if you're looking for a service online and find two similar providers, one with 38 likes and another with 3,000, the latter one's gonna' stand out - but with the practice of buying followers and likes so widespread, it'd be great to also have a way to work out who's telling the truth, right? Here's a couple of ways to work out if they're telling you the fibs.
How to work out is someone's Twitter followers are fake
Twitter is the open network, the one where people go to broadcast their thoughts and voice their opinions on the happenings of the world. As such, the biggest advantage of Twitter is that most of their data is publicly accessible, which makes it easier to work out what brands are doing, what strategies their employing - and also, whether they're faking. It's actually pretty easy to spot on Twitter, even without any significant investigation.
When looking through Twitter, it's not uncommon for a celebrity to have a follower to following ratio that looks something like this:
Gotye's not a prolific tweeter, and as such, he's not following a heap of people. But he's Gotye, he's a world-renowned musician, and his fans are keen to hear whatever it is he has to say - hence, despite him not following back many folk, he still has 414,000 followers. That makes sense for a public figure with a large fan base, but when you come across a non-public figure, someone you've never heard of, with a similar follower/following ratio, that's a pretty clear indicator that something's amiss.
There are a couple of options for testing this on Twitter - Status People's 'Fake Follower Check' is one, Social Bakers, too, has a free fake followers test you can use - but my favourite is Twitter Audit, it's free, quick and very easy to use. The difference between each of these, and why I prefer TwitterAudit, is the number of records (profiles) each app checks to get an indication of how many fake followers the user has.
Of course, the accuracy of each is relative to the amount of followers the subject has - the percentage of followers you're testing decreases in-line with increases in follower count - but generally this data has been found to be indicative, when compared with tests on a more comprehensive scale.
To conduct a Twitter Audit, you just enter the handle you want to check, sign-in with your Twitter credentials, and away you go. How the test works is, it takes a look at that random sample of up to 5,000 of the person's followers and it looks at a range of factors for each - number of tweets sent, date of last tweet, follower/following ratio, etc. From this, the system determines which of those tested profiles are likely fake, then gives you a percentage and pie chart based on those findings:
There is, of course, a margin of error, but it's normally a fairly accurate indicator, particularly when analysing profiles with less than 5k followers.
To clarify and confirm the data further, you can conduct a manual check - paid tools like Followerwonk or Socialbro provide in-depth reports on follower growth over time. If you look up a profile and find any big drops or jumps in their follower numbers, like this:
Pretty safe to assume those followers didn't all randomly switch off in the same week (unless, of course, there was an offending tweet or similar logical connection).
Using the available apps, it's pretty easy to work out Twitter fakes. Twitter's always working to eliminate illegitimate profiles, so we might one day see an Instagram style purge with a heap of celebrities taking hits. But till then, if you ever need confirmation, just run 'em through a Twitter Audit, then sit back and scoff at their vanity.
How to work out is someone's Facebook likes are fake
Facebook fakers are a little harder to pin down. Unlike Twitter, most of Facebook's data is locked up or hidden behind privacy settings, making it a bit more difficult to determine, definitively, if someone's cheating. There's a few ways to go about it and while none of them will provide as clear a result as the Twitter audit options, they will give you some idea as to what's going on with any given page.
Find out where their fans are from
So, let's say that the Facebook page you're looking at is a local business - they work within your local region, they not a subsidiary of a larger international corporation - the people they work with are, logically, going to be based in the local area. The people who sell Facebook likes tend to be from third world nations - as noted in this piece from Copyblogger. Most of the fake likes you'll come across originate from Bangladesh, India, Egypt, Pakistan, Afghanistan, Syria and Indonesia. Now, that's not to point the finger and say all of the 'click farms' in the world are based in these regions, but if our local business has a heap of likers from these nations, that's a likely indication that their faking it. So how do work this out?
Facebook's graph search enables you to search for a heap of different parameters. The one we can use in this case is:
You insert the name of the business page at the end and it'll give you a display of all the hometowns of people who like that page. The problem with this is that Graph Search results are sorted based on affinity - how they're connected to you - not by total number, so you can't necessarily determine where the majority of this page's likes come from, but if it's a local business and they have a range of the above mentioned nations among the hometowns of their followers, you may have reason to question why they're showing up there.
Extra note: In this piece by Miguel Bravo, Bravo also suggests that the results of Graph Search queries like:
'Pages likes by people who like [insert page name]'
'Countries of people who like [insert page name]'
'Languages of people who like [insert page name]'
Can also produce telling results (and they definitely do in the example he's provided).
Check their interaction versus their Likes
This is a more tenuous linkage, but it can provide some insight. So, if the page you're looking at has 3,000 likes, you'd expect them to have a reasonable level of interaction on their posts, some discussion about their brand, right? You can do a quick assessment of their posts to see what sort of engagement they're getting on each - fake profiles are not going to interact with posts, so if they've got a crazy amount of page likes but are getting no action on their updates, they may have bought likes. Or they're not very good at understanding their audience.
By clicking on the actual 'Like' number on the page, you get a graph like this:
Now, dependent on other factors, this could be telling - a huge jump in likes on any given day indicates either a really popular post or promotion - or that the page has bought likes. You'll only be able to determine this by cross-checking the data against the posts. The other metric to consider is 'People Talking About This'. In this case, I'd be a little suspect, given they have 3,700 page likes - including a big boost in likes in the last week - yet only one person 'talking about this'. Again, these are not definitive measures - they can often end up being fuel for your own conspiracy theories, where you're really seeing what you want to see, but having a look at the numbers can be revealing on a page that's clearly purchased fake likes.
Extra tip: Fake profiles tend to have no profile image, or odd-looking, copied images - this is another element to check to further your investigation.
Really find out where their fans are from
If you're really serious about finding Facebook fakers, paid app Fanpage Karma will give you a breakdown of the location of any page's likers.
This is one of the clearest indicators you can use to determine if the page has purchased likes - if the top countries are nations where the brand doesn't even operate, that's a fairly large red flag waving in your face.
On one hand, it's frustrating that there's not an easier way to determine Facebook fakers, as there is with Twitter, but on the other hand, it doesn't really matter either way - if they've purchased fake likes, there's not a heap you can do. I mean, you could, theoretically, go through their list of fans and report each fake profile one-by-one (which you can also do on Twitter) but obviously, that's pretty time consuming and with Facebook already dealing with thousands of reports per hour, it's hard to know if those efforts will actually cause any effect - that, and the fact that some like sellers offer a 'guarantee', where they'll replace removed spam accounts, lessens the potential impact of conducting your own faker crackdown. The ongoing updates to Facebook's news feed algorithm mean that purchasing likes will hurt pages more than help in the end, and Facebook is always working to eliminate fakes where they can. While a higher number of likes is better looking, as with most measurements in social, it's only one part of the larger picture, one indicator of potential success. You might have ten total likes and that could be more effective than a thousand, if those ten fans are engaged, paying clients, responsive to your messaging.
Quality Vs Quantity
And this is the key element in the popularity contest - the metrics only tell a part of the story. While I can understand why businesses might consider boosting their numbers, metrics are only one element of the social marketing puzzle. What's more, fake likes and followers hurt the core product of social platforms - there's already been questions about Twitter's actual user numbers with reports suggesting that 9% of profiles are fake. That sort of speculation hurts their brand sentiment and turns off potential investors - the fake profile industry is bad for social media business, and you best believe the platforms are doing all they can to identify and eradicate imposter accounts. As with Instagram, at any time you could see a similar cull on any platform - buying popularity could end up very embarrassing if you get caught out.
Any measurement is an indicator - Likes, followers, Klout, Kred - each, in itself, is something to consider, but the only way to confirm the true social credentials of a person or brand is to investigate them yourself. Look at their posts, their content, assess what they're doing. There may be a logical reason why their numbers are the way they are. Or there may not. 'Influence' is relative - conducting your own analysis will show you whose earned it and whose bought something resembling what influence should be.
