Most of us had a question about the Instagram algorithm / algorithms , which is currently the 4th social media platform with the most users.
“How Does the Instagram Algorithm Work?” With a real video published by Adam Mosseri (CEO of Instagram) in June 2021, we found answers to our questions about the logic of this algorithm.
Massori explained the reason why different algorithms are used instead of the chronological order of the main page or the stories as follows;
Looking at the data in 2016, they noticed that 70% or more of the content on Instagram could not be viewed. Also, almost half of his close friends’ posts were overlooked due to homepage rankings. For these reasons, they started to use “ranking”.
They tried to determine the “interest level” of the content that they were likely to see by converting the behavior of the users in the application into data. In this way, they aimed to increase the time they spend in the application by showing the posts that they are more likely to be interested in. Another important reason for including these algorithms is to eliminate the “waste of time” feeling in people. Because users use their time more effectively by seeing their close friends and loved ones more and accessing the content they are interested in more easily.
Instagram algorithm: Home Page and Stories
In Instagram algorithm rank the posts, the first step is to look at the stories you haven’t seen yet and your interest in them. In order to understand this level of interest, the data about the user, the people who sent the stories and the posts that the user liked in the past are looked at. With the help of this data, a “score” is obtained that shows the user’s interest in these posts.
When calculating this score, questions such as whether this post is a video or a photo, and what its content is. Secondly, information about the user who shared this content is discussed. For example, is the person who posted the post your friend, how often does he post, etc. Thirdly, information about the user, that is, the behavior of the user while using the application, is taken into account. For example, data such as whether your likes are predominantly videos or photos are used. The last factor used in calculating this score is the interaction between the user and the person who sent the post. For example, do you often comment on the post owner’s photos, do you like the posts he/she posts?
Considering this data, they ensure to increase the user’s experience in the application and to highlight the people they care about on their homepage. Since human psychology is complex and not fully known, convergences are often used in these calculations. Using this data, an attempt is made to predict the user’s movements. Instagram algorithm cares of 5 things.
The 5 most important criteria for making these predictions are:
- Time spent in post
- The probability that the user likes the post
- Possibility to comment
- Possibility of saving
- The probability of clicking on the profile
Instagram “discover” is the part that works independently of the people. We follow and is more about what we’re interested in. Again, in order to find these topics and posts of interest, Instagram applies a filtering method called “collaborative filtering”. In this method, the posts that the user has liked in the past are examined. With the help of the data of other users who liked the same post.
Similar content that the user may like is presented in the Discover section. Instagram algorithm :
4 criteria to be considered when choosing the posts to be suggested in the Discover section,
- Data about the post (likes, comments, topics, etc.)
- Past interaction with the account that owns the post
- User’s own activity (watching too many recipe videos etc.)
- Frequency of the post owner
- Based on this information, the number of likes, saves and shares the post will have is estimated. In this way, an idea of the user’s interest in that post is obtained and the discovery is organized accordingly.
Ranking of Reels
Instagram algorithm, as in the discover section, the posts you come across in Reels are brought before you according to the topics you are interested in, rather than the accounts you follow.
The signal taken into account for the ranking of the video to be presented to the user in Reels,
- User activity (what kind of videos do you watch, like)
- Interaction between the user and the post owner (whether he watched the video before, interaction with his previous posts)
- Details about reels (when it was shared, how many minutes of video, etc.)
- Data belonging to the person sharing Reel
The content that will appear in the application also changes, taking into account the time it takes for users to comment. Save the post, like and spend time on the post.
Although Adam Mosseri stated that algorithms are not 100% effective and we may encounter errors from time to time. With this ranking system, posts that attract the attention of users are highlighted. And in this way, it is aimed that users spend more time in the application.
Looking at the data, we can easily see that as a result of the introduction of Reels and the support of Discover with an algorithm, the average usage time targeted by Adam Mosseri has increased. Although the use of algorithms is very well implemented in Reels and Discover, it can cause important oversight because the relevance ranking in stories and posts cannot be created exactly. This leads to negative comments and criticisms of some users.
Click this link to analyze your social media for free here
Follow me from Instagram here