Facebook announced a few specifics regarding changes coming to the news feed algorithm (EdgeRank). Overall the algorithm still works mostly the same, with the exception of a few new improvements. We’ll explain each change, and what you can do about it to stay on top of your strategy.
Facebook held a small gathering to explain their latest changes to the news feed. A few key take aways:
- There is still an algorithm determining the news feed
- New Story Bumping Signal
- New Last Actor Signal
- Future New Chronological by Actor Signal
- There will not be a purely chronological change
Story Bumping Signal
Story Bumping is a new Organic Ranking factor. Each visit, Facebook looks at all of the new content and creates corresponding scores. A Post decays on average after about 3 hours because of Time Decay. Therefore most people’s news feeds lack old content. With Story Bumping, objects with high EdgeRank (and assumably now very low values of Time Decay) can be bumped into your feed if you “missed the update”. This update to the EdgeRank algorithm ultimately decreases the effect of Time Decay.
Facebook explained Story Bumping as:
“When users load up Facebook in the morning, a finite number of stories are shown on screen. They can scroll to view additional stories, but inevitably users leave some unread before clicking elsewhere. Story bumping ensures that when the user logs-in to Facebook later in the day, some of these unread stories are brought to the top of their News Feed alongside whatever has been posted since they last accessed the social network.”
To accomplish this, Facebook is recording which posts are being seen by which users. This recorded history helps aide which stories to bump.
Facebook has A/B tested this and found an increase in engagement. Stories read also increased significantly from 57% to 70%. Story Bumping has been rolled out to all users on the web, but has not rolled out to mobile.
How To Take Advantage of Story Bumping
We believe that Post Lifetimes will now have a larger variance. Posts that are inherently good will most likely be able to pick up engagement well past the typical 3 hour average we’ve seen in the past. However, posts that are inherently bad will most likely die a quicker news feed death.
It will now be more important to know how quickly a post is likely to die. If the post seems destined for a quick death, the next window of opportunity is much sooner than you may typically expect. We’ll be studying this in the future.
Last Actor Signal
Facebook keeps track of who a user has made the last 50 interactions with. People & Pages that have been interacted with these last 50 interactions, will receive a small bump in value (Affinity). Ultimately, these interactions are now having a stronger influence into the Affinity component.
How To Take Advantage of Last Actor?
For Pages, it is ideal to be one of those last 50 interactions to receive that bump. For some brands, this will simply mean delivering excellent content at the right intervals (optimal post frequencies). Some brands will be able to significantly increase their post frequencies, and this may aide the likelihood of being included as one of the Last Actor’s actions.
Chronological by Actor Signal
Facebook has not rolled this out. Facebook is considering rolling it out in the near future, once they’ve optimized a few things. Initial results have not been as successful as they would have liked.
Facebook attempted to make a Chronological By Actor tweak to the feed. This will help with understanding real time content. This change essentially removes Time Decay from the algorithm, while increasing the value of the Last Actor signal, therefore giving the user the ability to view interesting content in a chronological listing.
How To Take Advantage of Chronological by Actor
There is nothing to do but to wait until this is rolled out. This may or may not affect Pages.
Is EdgeRank Dead?
Facebook does not have an official name for the algorithm. Facebook no longer uses the word internally, however they apparently refer to it as the “algorithm for the news feed”. Not quite as catchy or easy off the tongue, as EdgeRank. EdgeRank still seems to be the most popular unofficial term around, and we’ll be the first to let you know if that changes.
Lars Backstrom, the engineering manager of News Feed rankings at Facebook, said that “the three primary EdgeRank factors — Affinity, Weight and Time Decay — are still important pieces of what is today a much more complicated News Feed ranking algorithm.”