A Twitter experiment on fake followers

4 minutes estimated reading time

In 2013, I worked with a number of companies and executives that has used services to inflate their Sina Weibo accounts standing with fake followers.  (In case you haven’t heard about it, Sina Weibo is a Chinese micro-blogging service). Often this was seen as a cheap way of getting an ego-boosting metric on their account or showing a positive delta relatively cheaply.

The idea of follower count as a measure of worth can be seen in real life. Look at the front cover of magazines and the value of a celebrity is often measured in the number of followers that they have.
Take this Elle magazine cover and interview with Kim Kardashian. Kardashian’s social media presence is as much a mark of status as a Hermès Birkin bag.
ingrid's birkin
However with Birkin bags going for as much as £55,000 online (holds head in hands and rocks slowly back-and-forth), getting fake followers on social platforms like Twitter might be a bit easier. If Twitter isn’t your thing, there are a number of suppliers of fake followers for Instagram, fake YouTube subscribers and fake Facebook likes too.

I want to reiterate this here, acquiring fake followers is a cross-platform issue, Twitter is just a handy canary in the coal mine. The reason for this is the amount of tools that are are available to measure Twitter as a platform and the prevalence of its use amongst media elites.

Looking at the variety of fake follower services, it seems to be a thriving business. A quick search on buy twitter followers cheap gave me over 31,100,000 results according to Google, and who am I to argue with their maths?

De Micheli and Stroppa estimated that fake accounts used as fake followers accounted for 4 per cent of the Twitter user base, whilst Twitter claimed that the number was 5% in the S-1 document it filed prior to its IPO. The estimates of fake accounts on Sina Weibo are thought to be as high as 30% of the user base – it is hard to tell because Chinese Weibo consumption tends to be largely passive using it as a news stream rather than a ‘social’ channel.

Politicians and celebrities have both been caught out using fake followers to bolster numbers (presumably to add credibility to their social presence).

So what are the benefits and how does it work?

I added 55,000 fake followers on Twitter, so that you don’t have to.

The process itself is really easy. The fake follower services generally accept payment by PayPal and have easy-to-use e-commerce services. The followers are generally delivered over two-to-five working days.

First I tried buying a big batch of followers, 50,000. The supplier delivered 56,000+ followers, but the number declines by about 100 followers over a week or so. This number seems to be pretty consistent. I can’t work out if this is just a common business practice or a part of Twitter’s ongoing conflict with fake accounts.

The purchase didn’t:

  • Move my Klout score
  • Improve the quality or number of organic followers that I received

I made a second purchase with a different vendor for 5,000 fake followers. This was delivered over five days, again an extra 10 per cent of followers were added on the top by the vendor and a similar decay pattern of about 5% of the followers occurred. This small increase had more of an effect on Klout causing a temporary bump in the score.  It has a more pronounced effect on Sysomos authority measure bumping my authority from 7 to 9 out of a possible 10.

It didn’t improve the volume or quality of followers that my account got organically.

With both vendors at least 5 per cent of the accounts that they used seemed to real people’s dormant accounts that had been co-opted into the fake follower game. There obviously seemed to be a market in taking over accounts that had been dormant for over 12 months. None of the fake follower accounts were set to private – this could factor into developing a heuristic for looking at fake follower accounts?

My overall conclusion on the fake follower business is that it almost purely about personal vanity rather than gaming a system. More related content here.

More information

Pay up and embrace Twitter’s fake followers | Marketing Week
Fake Twitter followers: An easy game, but not worth the risk | The Next Web
How the market in ‘fake’ Twitter followers works | Yahoo! News
Rihanna Loses 1.2 Million Instagram Followers After Spambot Purge | Gigwise
Instagram makes teens and celebrities angry by killing millions of spambots | The Verge
Twitter and the underground market by Carlo De Micheli & Andrea Stroppa at 11th Nexa Lunch Seminar, Turin, Italy (May 22, 2013) – PDF
Inside a Twitter Robot Factory | WSJ
Twitter Admits 5% Of Its ‘Users’ Are Fake | Business Insider
I Bought 10,000 Fake Twitter Followers. Why Didn’t Klout, Kred (or Others) Notice? | Ignite Social Media