For this project, we wanted to help people evolve their social graph on Instagram to always reflect their current close friends and accounts they care about as their social circles and interests change over time.
However, many people have accumulated large graphs on Instagram over years. It would take a lot of work to go through their following list to manage who they follow. Even for people who were aware of the need to do so, they perceived the effort and didn't engage. So the goal was to make it easier for people to meaningfully trim their graphs.
Here are the approaches I took:
-
Do the hard work for people.
-
Use signals of intent to encourage more and deeper engagement
-
Create a logical and reliable structure.
I mainly looked at two types of trimming - active trimming (going to self following list to do "spring cleaning" once in a while) and serendipitous trimming (unfollowing accounts after seeing their content in feed or stories), and wanted to improve both.
Self Following List
This was one of the main surface I looked at. The goal was to turn this list into the go-to surface for people to mange their graphs. After rounds of explorations and iterations, I landed on this smart groups idea, where we could surface accounts people might want to unfollow in shorter and snackable lists.
Unfollow Chaining
Besides making it easier for people who do active trimming, I also want to help the majority of users who only engage in serendipitous trimming. I came up with idea, unfollow chaining, which surfaces similar accounts that the user follows after unfollowing an interest account.
After it launched in 2020, it got great feedback and tons of press coverage - The Verge, TechCrunch, TNW, Refinery29, etc.