EdgeRank died a quiet death more than two years ago. Surprised? So were we when we learned of its untimely death on August 16 from a report in Marketing Land.
"EdgeRank" was the name given to Facebook's news feed algorithm years ago, and many still refer to it as EdgeRank today. While the name has been officially declared dead, and the algorithm improved, the framework is still alive and well.
"It's changed a lot over the years," said Jeff Widman, director of research at Unified and author of EdgeRank.net. "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, it's absolutely true that the algorithm has become more complicated."
Widman said the algorithm can be summed up easily by understanding this concept: Actions you take on Facebook are going to impact what you and your friends see in the news feed.
When Facebook originally revealed the EdgeRank algorithm, it was based on three weighting concepts: affinity score, edge weight and time decay. From EdgeRank.net:
- Affinity score: Facebook calculates affinity score by looking at explicit actions that users take such as the strength of the action, how close the person who took the action was to you and how long ago they took the action.
- Edge weight: Every action that a user takes creates an edge, and each of those edges, except for clicks, creates a potential story. Each category of edges has a different default weight.
- Time decay: As a story gets older, it loses points because it's "old news." EdgeRank is a running score – not a one-time score. When a user logs into Facebook, their newsfeed is populated with edges that have the highest score at that very moment in time.
Widman said that even though the news feed algorithm may have a new name and more sophisticated way of ranking stories (Facebook now says there are "more than 100,000 factors"), the algorithmic factors still fall into the three buckets of affinity score, edge weight, and time decay.
Image Credit: EdgeRank.net
"Whether or not they have 100,000 factors or just 10, if you think about the design of the algorithm, every single factor in the news feed algorithm is still going to fall under those three categories," Widman said.
New Features, Same Foundation and Goals
Regardless of how the algorithm evolves, Widman said the intention is the same. "Facebook's high-level goal in all of this is to show content to users that they find interesting, so that they will keep coming back to the site."
As marketers, knowing what users find interesting is important to the experience they create for brands on Facebook. To aid in that, Facebook recently announced it would make public announcements about updates to its algorithm – similar to how Google does.
Facebook made good on its promise when it announced publically a new way it was going to rank stories in the news feed called "story bumping." With story bumping, Facebook said popular page posts could have a higher chance of being shown even if they are a few hours old (referencing the time decay element) by going to the top of the news feed if the stories were still receiving a lot of likes and comments (referencing affinity score and edge weight elements).
This story prioritization is a classic example of how Facebook is still utilizing the concepts of what was formerly known as EdgeRank, but in a new way.
However, Facebook announced last Friday that it would be rolling out an algorithm update that factors some new concepts into the news feed aimed at filtering out low-quality content.
Reminiscent of Google's methodology for determining quality, Facebook turned to its users in a survey and asked:
- Is this timely and relevant content?
- Is this content from a source you would trust?
- Would you share it with friends or recommend it to others?
- Is the content genuinely interesting to you or is it trying to game News Feed distribution? (e.g., asking for people to like the content)
- Would you call this a low quality post or meme?
- Would you complain about seeing this content in your News Feed?
From Facebook's announcement:
We used the results of this survey to build a new machine learning system to detect content defined as high quality. The system uses over a thousand different factors, such as how frequently content from a certain Page is reported as low quality (e.g., hiding a Page post), how complete the Page profile is, and whether the fan base for a particular Page overlaps with the fan base of other known high quality Pages.
What's in a Name?
Web marketers have a love/hate relationship with the way things are named. Because technology and tactics evolve so quickly in the industry, a name that once made sense can become the center of a heated debate as the concept grows and changes. Take the term "SEO," for example.
So if the Facebook algorithm has evolved, does that mean we should stop using the dated "EdgeRank" name? Dave Donohue, vice president of marketing communications at Unified said if Facebook isn't using it, we shouldn't be either.
"We don't use the term, our customers don't use the term, Facebook doesn't use the term," Donohue said. "Conceptually, the EdgeRank algorithm is alive, but the word 'EdgeRank' is dead."
Widman said that regardless of the name used, the more important thing is understanding the evolving algorithm itself. "If you're not successful in the news feed, you're not going to be successful on Facebook."
And as you're sifting through the factors in the algorithm and the tools that help you understand it better, Widman said it's very important to keep in mind that there isn't a real way to extract data on things like your story's true position in the news feed. Similar to Google, Facebook serves up its results based on many factors that change from user to user.
While getting rid of the dated term that marketers have embraced for years may take some time, a heavier push by Facebook to keep brands updated on the news feed algorithm is certain to help shift the mindset to not only adopt a new name, but also new ways to keep users engaged.