Like, Comment, Repin: User Interaction on Pinterest

TL;DR

This study analyzes Pinterest's user interaction network, revealing that most activity comes from non-followers and that followers who are active perform twice as many actions, highlighting the difference between interaction and follower networks.

Contribution

We provide the first large-scale analysis of Pinterest's interaction graph, showing its distinct structure from the follower network and quantifying user engagement patterns.

Findings

Only 12.3% of followers interact with a user's pins.

Over 70% of activity is from non-followers.

Active followers perform twice as many actions as non-followers.

Abstract

We present the results of a study of the Pinterest activity graph. Pinterest is a relatively new and extremely popular content-based social network. Building on a body of work showing that the hidden network whose edges are actual interactions between users is more informative about social relationships that the follower-following network, we study the activity graph composed of links formed by liking, commenting, and repinning, and show that it is very different from the follow network. We collect data about 14 million pins, 7 million repins, 1.6 million likes, and several hundred thousand users and report interesting results about social activity on Pinterest. In particular, we discover that only 12.3% of a user's followers interact with their pins, and over 70% of activity on each user's boards is done by non-followers, but on average, followers who are active perform twice as many actions as non-followers.