Instagram explains how recommendations work
The company has published this new blog post in an attempt to clear “misconceptions” about the oh-so-infamous algorithm it’s using. Instagram’s head Adam Mosseri dives deep into the explanations about the rankings.
First off, he notes that there isn’t a single algorithm that oversees what people do on the app. He explains there are multiple algorithms and tanking systems, and they are responsible for different aspects of the app, like Explore, Reels, Stories, and Search. He adds that each algorithm uses a plethora of signals to determine how content is shown to each user.
(Image Source – Instagram)
For instance, the posts you see in your feed are determined by your past activity, as well as the interactions you have with the person who made the post. Stories are determined by a similar tactic: for example how likely are you to be friends or family with the person.Meanwhile, recommendations in Explore follow a slightly different approach. Those are based on posts you’ve liked, saved, shared, and commented on in the past, but it’s more likely these posts come from people you haven’t interacted with.
All of this sounds pretty straightforward. Mosseri also goes on to address another aspect of Instagram that has been bugging some creators. The so-called “Shadowbanning”.
Mosseri also addresses Shadowbanning
First off, for those of you who may not know, Shadowbanning is not an official term and is been used by creators who feel their accounts or posts have suddenly gotten a more limited exposure, or are hidden, and that is without any warning or explanation on the part of the social media. Mosseri says that the company is working on increasing transparency if a creator’s content or account is blocked from the app’s recommendations. He says that the app’s account status feature can alert users if their content or their account is considered not eligible for recommendations. You will also have the option to appeal.
On top of that, Mosseri notes that the app is testing new notifications that will be aimed to help creators understand when the reach of their reel may be limited due to a watermark (a watermark appears when a short vid is taken straight out of TikTok and posted onto Reels).