![[Video tutorial] Understanding EdgeRank, a comprehensive guide for 2016.](https://fanpagecash.org/wp-content/uploads/2026/10/fpc-video-tutorial-understanding-edgerank-a-comprehensive-guide-8183-1024x538.jpg)
From the archive. First published in February 2016. Rewritten for clarity, with a note on what happened afterwards.
EdgeRank was the name marketers gave to Facebook’s early feed algorithm, usually described as three factors: affinity between the viewer and the poster, a weight for the type of interaction, and time decay favouring newer posts.
It is worth stating clearly, because the term still circulates: Facebook stopped using that system around 2011. It said so publicly in 2013, noting it had dropped the name roughly two and a half years earlier. EdgeRank was never an official product name in the way marketers used it, and guides explaining how to optimise for it have been describing something that does not exist for fifteen years.
What replaced it
Machine learning across an enormous number of signals. Meta describes a four stage process: gathering available inventory, reading signals, running predictions, then producing a relevance score. More than a hundred prediction models are involved. Reported signal counts vary wildly between sources, from a thousand to a hundred thousand, and Facebook has never published a definitive figure.
The part that actually matters today
Ranking now runs on two different signal sets depending on whether content is connected or unconnected to the viewer.
- Connected content, meaning people and pages you follow, is ranked partly on your past interaction with that poster.
- Unconnected content, meaning recommendations from accounts you do not follow, leans on post type, watch time, what you have asked to see more of, and topics you have engaged with recently.
Since roughly a third of organic feed views in the United States are recommendations, the second set is where most of the opportunity sits, and it has nothing to do with your followers.
The practical translation
Stop optimising for a relationship with your existing audience and start optimising for recommendability: original content, native format, watch time, and no signals that cost you recommendation eligibility. The checklist is in our guide to organic reach.
Thinking of selling?
We give free valuations with no obligation. Send us the page URL and a representative replies with a figure, usually within one business day. Looking to buy instead? Register as a buyer.
