Misuse on social media – the case of D.K.

Soon after the rise of the “popularity” sites, about which was written in the first text,  marketing companies and experts, in order to prevent to be defrauded and to pay fake influencers for marketing services, have developed their own tools for „quality“ and „authenticity“ check of followers and likes. So that today with a small amount of money and using websites such as fakelikes.info you may check “the quality” of Instagram and Facebook influencers, or the number of their real and fake profiles that followers and likes them, compared to the number of profiles artificially created and purchased through different online shops.

Currently, for us, it is maybe the case of a businessman, Dusko Knezevic – who opted for Instagram as a political communication channel with the public – the most interesting, particularly for two reasons.

First is the anomaly noticed in December 2018 when Dusko Knezevic on the photo posted on December 14 had 156 likes, while on the photo posted just a day after, i.e. on December 15, had 2,130 likes. For us who deal with social networks, it was clear that such leaps in the number of likes do not happen overnight, so that was the “alarm” for more detailed research.

The second reason is a typical example of bot profiles misuse in order to make a fake performance of support and popularity, when actually it was a common purchase of bot profiles and journalistic irresponsibility to report on unverified data, as was the case with portals slobodacg.me and in4s.net.

For the reason of the topic of this Magazine issue, we did the quality analysis of Instagram profile @dkatlas.

Let’s begin with the engagement rate (ER), which represents the total number of people commenting on or liking the posts compared to the total number of followers. In the case of this profile, it is about 4.97%. When bot profiles commenting on and liking the content are removed, it remains a so-called real engagement rate (real ER) which amounts to only 0.5% and represents comments and likes coming from real accounts, i.e. people.

When it comes to the followers, at the moment of the analysis, there were 60,772 followers; however, between 5,100 and 11,300 followers represent actually real accounts.

Likewise, out of the average 2,960 likes per post on the last 12 posts, a total of 2,667 (90%) likes constitute likes from bot profiles, i.e. fake likes.

Compared to the accounts with a similar number of followers, the @dkatlas account has 4.9 times lower engagement rate. As is stated on the site fakelikes.info, if an account has at least 2 times lower engagement rate than the accounts with the similar number of followers, then it is most probably about fake accounts and followers.

The post from May 24 – “DK AT THE BRITISH PARLIAMENT at the panel on Western Balkans” – is dominated by fake comments, as many as 93.1%, what may be seen from the mere overview of the “comments” section below the photo.

Almost the same situation may be seen on other posts of the analyzed profile, whether it is about likes or comments.

The analyzed profile of Dusko Knezevic is only one example of the possibility to buy followers, the number of likes, the number of views of video contents. All this has opened the additional space for misuse on social media.

Fake followers and likes have always been explicitly contrary to Instagram’s Regulations and Terms and Conditions. This platform has started its war against fake activity long ago, using tools which instantly identify accounts using popularity purchase sites, and removing comments and likes gotten in such way.