Twitter's IPO and the curious phenomenon of paid followers on Twitter
When we get approached for quote requests we always do some due diligence on the company making the request. This generally takes the form of checking their recent accounts filed at Companies House and researching their website and wider online presence. There have been several occasions recently where I have checked out a company's website, seen they have a Twitter account, browsed to it on Twitter and simply thought 'Wow!'.
What I'm talking about is the standout number of 'followers' that the Twitter accounts of some small businesses have. I'm not going to name specific examples in this article but when I see that a company has a very modest 'book' value according to their company accounts, plus I've never heard of them, I generally expect them to have a low number of Twitter follows. This applies even if their Twitter account's content is good. Like all small business owners I am only too well aware that raising your public profile is a very difficult task indeed. So when I see their Twitter account has thousands of followers it raises a mental 'red flag' that something doesn't stack up.
After running into this situation a few times I decided to dig a little deeper via the obvious route of Googling 'buy Twitter followers'. Lo and behold there are a plethora of dubious online services offering to provide you with a 'guaranteed' number of Twitter followers, typically within a 3 day period, for a small payment. There is even a comparison site that helps you decide which paid Twitter follower service to use! Established sites offering this service include:
Typically $12 will buy you 1000 followers and if you're feeling flush you can pay all the way up to $400 for 100,000 followers. That's a lot of fake Twitter accounts! To be offering this kind of scale of followers then these follower Twitter accounts must be programmatically created and controlled, a bit like the 'bot nets' you sometimes hear about in the large scale internet attacks that take down the servers of government agencies and big corporates.
Do paid for Twitter follower services work?
Well there's only one way to find out. I'm not cynical enough to use this to boost our own fledgling Twitter account. Instead I created a dummy Twitter account as a test case: https://twitter.com/Not_ZodiacMedia. A screenshot taken shortly after I had created the account (13:42 BST 18th October 2013) showed the account had a single follower:
Immediately after taking this screenshot I ordered up 1000 Twitter followers for the account at a cost of $10.80. When I checked the account the next day (14:57 BST 19th October 2013) the account had a very impressive 1179 followers.
So buying fake Twitter followers definitely works!
Interestingly the service I used to buy Twitter followers with must have accidentally doubled my order as when I checked again one day later (09:59 BST 20th October 2013) the account's followers at approximately doubled to 2357.
A quick browse through the profiles of some of these new followers indicates that they won't be reading any of my Tweets nor sharing them with their followers by re-tweeting them. The accounts seem to be a 'Frankenstein' assembly of parts of other Twitter profiles as is shown by the selection of screenshots below.
This is exactly as I had expected so I don't feel short changed by my paid for Twitter followers. The big benefit I have bought is that my account now has a much more impressive number of followers when someone comes to view it in the future. This gives the much sought after impression of popularity and success. If this were our actual Twitter account then it would help add to our credibility and persuade someone to either follow us on Twitter or, even better, do business with us.
Can you tell if a Twitter account has fake followers?
Now this is almost getting fun! Websites exist that purport to be able to give an indication as to how many of a Twitter account's followers are 'fake', 'inactive' or 'good' (see: http://fakers.statuspeople.com/Fakers for example) . Admittedly this is only ballpark stuff (http://fakers.statuspeople.com/Fakers/FindOutMore/) but I believe that logically it would be possible to place an account's followers into these three categories on the basis of how complete their Twitter profile is (e.g. real name, photo, profile summary etc) and how many Tweets and followers they have.
To make this more fun I'm going to run through an 'Indicated fake followers competition' for some leading UK political figures, plus some other 'yardstick' Twitter accounts.
Step forward contestants:
|David Cameron||Prime Minister and Leader of the Conservative party||487k|
|Ed Miliband||Leader of the Labour party||259k|
|Nick Clegg||Deputy Prime Minister and Leader of the Liberal Democrat party||134k|
|Boris Johnson||Mayor of London (and possible future leader of the Conservative party)||798k|
|Vince Cable||Business Secretary (and possible future leader of the Liberal Democrat party)||46k|
|Private Eye Magazine||Satirical news and current affairs magazine||284k|
|Wayne Rooney||England and Manchester United football player||7.36m|
|World First||FX brokerage (and my previous employer)||9k|
|Not Zodiac Media||Our fake Twitter account||2k|
And, drum roll please as we announce our results!
|Name||Fake (%)||Inactive (%)||Good (%)|
|Private Eye Magazine||19||45||36|
|Not Zodiac Media||100||0||0|
Of course these results don't really mean anything, they're just a bit of fun. We are making the assumption that the source of the data is accurate which it's unlikely to be: http://fakers.statuspeople.com/Fakers/FindOutMore/. Besides, even if a Twitter user has indicated 'fake' followers they may not have paid to obtain them. Indeed it would be possible to smear a Twitter user by buying fake followers for their account and waiting until somebody else discovers the fraud.
The results do indicate however that 'fake' and 'inactive' users appear prevalent in Twitter.
What do fake and inactive Twitter users mean for Twitter's IPO
Short answer – not much. Twitter still provides a useful service to a lot of people and it's unlikely that initial investors will be aware of how prolific fake and inactive users are on Twitter. It's not possible to obtain detailed and accurate data on inactive and fake Twitter accounts unless Twitter were to release this data themselves. It's not in Twitter's interest to do this, although they did cover it to an extent in their recent IPO S-1 filing at the US Securities and Exchange Commission.
In the "Prospectus Summary" of Twitter's IPO document Twitter states:
"People are at the heart of Twitter. We have already achieved significant global scale, and we continue to grow. We have more than 215 million monthly active users, or MAUs, and more than 100 million daily active users, spanning nearly every country."Source: Twitter S-1 filing
The definition of a Monthly Active User (MAU) is given in the IPO document as:
"Monthly Active Users (MAUs). We define MAUs as Twitter users who logged in and accessed Twitter through our website, mobile website, desktop or mobile applications, SMS or registered third-party applications or websites in the 30-day period ending on the date of measurement. Average MAUs for a period represent the average of the MAUs at the end of each month during the period."Source: Twitter S-1 filing
Further on in the IPO document in the section entitled "Risk Factors" is the following:
"We rely on assumptions and estimates to calculate certain of our key metrics, and real or perceived inaccuracies in such metrics may harm our reputation and negatively affect our business.
The numbers of our active users and timeline views are calculated using internal company data that has not been independently verified. While these numbers are based on what we believe to be reasonable calculations for the applicable period of measurement, there are inherent challenges in measuring usage and user engagement across our large user base around the world. For example, there are a number of false or spam accounts in existence on our platform. We currently estimate that false or spam accounts represent less than 5% of our MAUs. However, this estimate is based on an internal review of a sample of accounts and we apply significant judgment in making this determination. As such, our estimation of false or spam accounts may not accurately represent the actual number of such accounts, and the actual number of false or spam accounts could be higher than we have currently estimated."Source: Twitter S-1 filing
So Twitter believe that 5% of their MAUs are fake, equating to approximately 11 million MAUs. This seems possible but only if the businesses offering fake followers are keeping their fake follower accounts at a constant number. If I buy 100,000 fake Twitter followers are 100,000 new Twitter accounts created? Also, although the number of MAUs is rising month on month it would be interesting to know what this would look like if spam accounts were stripped out. As Twitter grows in importance then the number of spam accounts is also likely to grow. Twitter may be cullling 'fake' accounts every month, but if the number of spam accounts is growing rapidly then this may distort the MAU figures.
Further research online indicates that I am not the only person to have raised concerns about this:
- The Wall Street Journal – 3rd October 2013 - One Big Doubt Hanging Over Twitter's IPO: Fake Accounts
- The Register – 4th October 2013 - The DARK HEART of the Twitter IPO: FAKE USERS
Those considering buying in to the Twitter IPO - Caveat Emptor!