The Facebook EdgeRank Algorithm

Facebook. Beyond its immense social networking power, its 500 million members, and the fact that its homepage is now the most frequented web page on the entire Internet, Facebook is a place where people share content; lots of content. Over 360 billion pieces of content are shared on Facebook every single year. That’s around 1 billion individual posts daily, over 694 thousand posts every minute and just over 11,500 posts every second (for those who are counting). Oh, and it’s increasing all the time.

From an Internet marketing perspective, spreading content virally through Facebook can be a great addition to almost any type of web marketing campaign. Everyone who’s used Facebook understands just how easily great content can be shared and circulated through people’s news feeds. But how exactly does Facebook determine which content should surface in people’s news feeds and which should not? Of all the thousands of posts made daily by your friends, how is Facebook able to figure out which one’s you’ll probably want to see versus the ones you won’t? Luckily for us, Facebook answered this very question at their 2010 F8 conference.

EdgeRank

Whenever you login to Facebook, what you see in your news feed is only a very small fraction of all the content being posted by your friends at any given time. In order to determine which items should appear within your news feed, Facebook grades all content posts according to what it thinks your level of interest with each of them is likely going to be. The mathematical formula that determines this for Facebook is known as Edgerank.

Some Terminology: Edges and Objects

Facebook uses the terms “object” and “edge” to help describe user’s interaction with content. An object simply refers to a post or status update made by a user; a piece of content. The term “edge” describes any user interaction with an object; a “like”, a comment, a tag, a “share” and the action of creating the content itself. (It’s probable that clicks and page views also qualify as edges, but I wasn’t able to definitely confirm this.)

The EdgeRank Formula

Here it is:

The variables:

U - the affinity score between viewing user and edge creator

W - weight score for this edge type (create, comment, like, tag etc)

D – time decay score based on how long ago the edge was created

The idea:

The higher the combined value of the affinity score, weight score, and time decay score, the more important an edge will be. Similarly, the more edges, and the higher their total scores, the higher an object’s EdgeRank. Objects with a higher EdgeRank appear more prominently within people’s news feeds on Facebook.

Affinity Score

Affinity is the value Facebook assigns to an object’s creator relative to other Facebook users. In plain English, the more a person comments, likes, or interacts with your posts, the higher their affinity to you, and the greater the likelihood that your posts will show up in their news feed.

For example, I’ve been following 6S Marketing social media strategist Sam Macmillan’s Facebook page because he recently became a finalist in, and then actually won, a local social media marketing contest. Since I’ve been viewing and interacting with Sam’s Facebook page relatively frequently, I have a high affinity to him on Facebook, and the content he posts regularly shows up within my home feed.

Weight Score

The weight score is a measure of the combined value of all an object’s edges. The more edges an object has the higher its weight score will be. Not all edges hold an equivalent weight on Facebook. So for example a comment likely holds more weight than a like or a simple click or a “like”.

Generally speaking, the more interactions a content post has on Facebook, the more prominent it will be in other people’s feeds. A post with 12 comments and 10 likes will appear in people’s news feeds more prominently than one with no comments and no likes.

Time Decay Score

Like Google, Facebook is concerned with the freshness of content. Newer content and interactions hold more importance and have a higher likelihood of being published in your news feed. The older the interactions are with an object, the lower their time decay score and less of ability they have to push content to the top of people’s news feeds.

What now?

While it’s interesting to have a better understanding of Facebook’s content ranking algorithm, much of this information reaffirms what we already know. It’s not surprising that the more people interact with your posts, and the more interactions there are with a post, the more likely that post will be to be visible in people’s feeds.

If you’re promoting a campaign on Facebook, be sure to share content that your audience will want to engage with. This doesn’t mean that everything you share has to be unique, it just means that the content you do share has to be able to draw the attention of your audience and then connect Facebook users with your message.

Many sophisticated online marketers allocate large budgets toward Facebook pay per click campaigns, but in contrast, every time a content item you post on Facebook surfaces in people’s feeds it’s a free opportunity to get your message out. Figuring out how to run content campaigns that actually get published within people’s news feeds can therefore save you money and get your message out.

Questions or comments? Don’t be shy, leave us a message in the comments section below. We love hearing your feedback.

Also be sure to check out 6S’s Facebook marketing services or social media strategy page for more information about our social media marketing services.