Last month, Facebook did away with ‘EdgeRank,’ the name given to its News Feed algorithm, and replaced it with a new, more complicated, machine-based learning algorithm. While the new algorithm takes into account far more factors when determining what a User will see in his/her News Feed, the algorithm’s frameworks function in the same way that EdgeRank did. Jeff Widman, author of EdgeRank.net, says, “At first, we called it the News Feed algorithm. Then Facebook started calling it EdgeRank. And then Facebook said it moved beyond the term EdgeRank. But no matter what you choose to call it, one thing remains the same… actions you take on Facebook are going to impact what you and your friends see in the news feed.”
Let’s first break down the algorithm’s basic, unchanged frameworks and then highlight the new factors that have been implemented.
What was Facebook’s EdgeRank?
EdgeRank was an algorithm developed by Facebook to govern what is displayed – and how high it’s displayed – on the News Feed. Facebook defined any action that happens within Facebook as an Edge. Examples of Edges would be status updates, comments, likes, and shares. Objects with the highest EdgeRank go to the top of the News Feed.
There were three factors that determined EdgeRank. The first factor, affinity, defined the one-way relationship between the User and the Edge. Affinity was built on repeated interactions a User would have with the Edge. For example, actions like commenting, liking, sharing and clicking all could influence the Edge’s affinity score. The second factor, weight, was a value system created to increase or decrease the value of certain actions within Facebook. In this weighting system, an action like commenting on a Post would have greater value than liking it. Essentially, the more time a User’s action would take, the greater the Edge’s weight. The final factor, time decay, defined how long an Edge has been alive. The older an Edge (e.g. a Post by a brand) the less valuable it becomes, which helps keep the News Feed fresh with relevant new content, as opposed to lingering old content.
How is the new algorithm different?
While the new News Feed algorithm is still using the same basic principles of EdgeRank, the science behind it is far more complicated. Facebook reports that there are more than 1,000 factors now determining the popularity and value of Posts. Content such as a status update, photo or video is scored based on a User’s relationship with the content (e.g. whether the User liked or commented on the Post.) Consequently, Users can positively affect the ranking by engaging with the content or negatively affect it by “hiding” the content from their feeds.
Two other noteworthy changes to the News Feed algorithm impact the way that content shows up on a User’s feed. The first factor, called story bumping, moves up older stories to the top of a User’s News Feed if he/she missed them. Previously, Facebook only posted the newest stories at the top of the feed to keep the freshest content at the top. While story bumping is not yet impacting Facebook’s mobile platform or advertisement placements, it has resulted in an 8% increase in likes, comments, shares on stories from Pages, according to a Forbes August 2013 report. Moreover, the number of stories read by Users increased from 53% to 70% in July 2013. The second factor, called the last actor effect, bumps content from a person or Page that you as the User have shown interest in over your last fifty content interactions.
What are the implications here?
Marketers need to be aware of Facebook’s changed News Feed algorithm, as it has become increasingly more difficult for a brand’s message to reach its fans. According to Facebook, on average, 1,500 pieces of content can appear in each person’s News Feed each day, yet due to space and time constraints, Facebook typically serves up only 300 pieces of this content a day. While the precise factors Facebook is using for this new algorithm remain a mystery, it’s evident that social media managers need to be extremely diligent when creating their content strategies, as only their most attention-getting posts will be seen by their fans.