An explanation on how content algorithm works on Instagram

In the current age, social media tends to become a crucial part of our everyday lives. The policies are developing again and again, and we left questioning how algorithms are working. To drop all the misconceptions, Instagram’s executive Adam Mosseri has given us an overview of the content promotion on multiple platforms in his recent blog.

As per the explanation provided in a recent blog post by Mosseri, other than looking over a single factor, the algorithm works on a complex web of factors, including reels, stories, and search, with a major part stemming from user-generated data.

Algorithm factors for stories and reels

As for stories, many factors are being ranked for engagement, including the update of accounts, interactions in DMs, and likes on a story. Other than that, it’s also evaluated over the relationship with the following accounts, either their family or friends.

However, in terms of reels, these factors differ slightly, as instead of pointing out interactions with specific accounts, Instagram looks for likes, saves, and shares on videos, including the type of content being displayed in the video. Along with that factor, factors like video resharing, completion rate, likes, and engagements with audio play a vital role for content observers.

How to tackle shadow banning

Shutting down an account or piece of content without a good reason or any proper explanation is known as shadow banning. Instagram has now addressed this issue and stated that they are actively working to improve transparency by introducing an “account status” function, following much rumours and conjecture. Users can appeal the decision using this function, which will also notify them if Instagram deems their content “ineligible” for recommendations.

While Instagram’s openness about its recommendation system is admirable, it’s critical to comprehend the complex nature of such algorithms because they depend on a vast amount of data and machine learning models. Therefore, it is impossible to give a clear definition. However, having knowledge of the underlying variables that influence recommendations will enable users to use the platform more skillfully.