Remember Facebook Questions? Checkins? Spotify listens? They had their time at the top of the news feed because Facebook favors new products, but their prominence diminished as they matured. This and competition for space between posts may explain why New York Times writer Nick Bilton gets fewer Likes on shares to his public followers than a year ago. The news feed is a bumpy ride.
Update: Now Facebook has issued a statement, admitting the drop in engagement seen by Bilton, others, and I is real, but that dismisses our accounts as anecdotes. Read my breakdown of Facebook's damage control: “Facebook Admits Likes Are Down For Some Celebrities' Posts, But Claims Up 34% For Most”]
According to Zuckerberg's Law Of Sharing, we post twice as much each year, but we're not doubling how long we spend reading our social streams. On Twitter's unfiltered feed, that means we read tweets from a shorter period each time we browse. The last 100 tweets may now come from the last hour, when perhaps it took those we follow two hours to conjure up as many quips and cat memes a year ago.
On Facebook's filtered feed, though, push comes to shove, as Hunter Walk mentions in his response to Bilton. Facebook shows a digest of the most “relevant” posts from the last few hours or since we last logged in. As we share more, the bar climbs, and only the posts with the most likes and comments and those from our closest friends show up.
Facebook also sidesteps its news feed sorting algorithm, unofficially known as EdgeRank, to inject certain pieces of content. For example, ads. Whether they're posts by Pages we Like that could have appeared but probably wouldn't make the cut, or non-social ads that are completely artificial, Facebook makes money by sticking them high in the news feed. The volume of advertising in the feed has increased dramatically this year, which Bilton says means “free posts will disappear from people's feeds as sponsored ads float to the top.”
It's a balancing act, and Facebook may be tipping too far towards advertisers' interests right now. I agree with Bilton. Facebook risks alienating users if the feed's meritocracy is poisoned with paid marketing and users feel like they're not seeing what they want.
But ads aren't the only thing Facebook gives an EdgeRank edge to. When it launches new products, it wants to raise awareness about them so people give them a try. So at first, Facebook pumps up the latest features with extra weight in the news feed algorithm.
For example, when Facebook first launched and re-launched Questions in 2010 and 2011, it flooded the news feed with queries and their answers. It did the same when it started letting people check in to local places or announce they were an organ donor. And most people remember the first few days after f8 in September 2011 when the new sidebar Ticker and the news feed were overrun with posts about what friends were listening to on Spotify. Eventually the presence of these posts were throttled back to more comfortable levels. In absolute terms, they were seen less over time, but mostly because they were purposefully shown too much at first.
The fluctuations don't cause much hubbub when the posts are just fun updates between friends. They cause a big stir when the hammer drops on third-party app developers, websites, and professional content producers.
Soon after f8, Facebook was frequently showing a “Recently Read Articles” box with links to news reader apps from the Wall Street Journal, and other outlets. They drove tons of traffic. But then Facebook stopped showing the box as much, and traffic plummeted. At first, BuzzFeed and others claimed users were ditching the annoying reader apps that auto-posted their news consumption. I explained and The Washington Post confirmed it was actually news feed changes that caused reader app traffic to plummet.
Facebook also meddles with the feed when it wants to box out competitors. Back in 2011, Twitter was getting a lot of love for being a powerful distribution channel for real-time news. By featuring reader apps in the feed, Facebook seemed intent on stealing some of Twitter's spotlight and winning the hearts of news outlets.
The Subscribe feature seems to have been chasing the same goal. Until its launch just before f8, individuals could only have 5,000 friends on Facebook. If they wanted more, they had to start a Page - a product designed for big brands, not journalists and micro-celebrities. These people were gaining huge followings on Twitter, and their content was drawing people to the land of 140 characters.
Facebook didn't like that one bit. So it launched Subscribe so individuals could use their same personal account but rack up as many followers as they wanted who would see their public posts. Facebook promoted the accounts of early Subscribe adopters by showing them in “People To Subscribe To” boxes in the site's sidebars. In just three months, my Facebook subscriber count surpassed that of my Twitter followship I'd spent several years building. Facebook was clearly trying to convince journalists like me and other public figures that it was more valuable than Twitter.
Now looking back, it seems Facebook may have been juicing the news feed prominence of posts to subscribers to the same end. At this point I have 160,000 Facebook subscribers, but I'm getting the same number or fewer Likes on my public posts now as when I had one-tenth as many subscribers. My Likes and comments haven't dropped as much as Bilton's but the fact that my engagement isn't directly correlated with my growing subscriber count implies that public posts are indeed getting throttled.
Posting to Facebook is free, but the cost is acceptance of Facebook's iron grip on the opaque news feed. Facebook's still a gift horse, but it's an inconsistent one. We might not want to saddle up and build a business completely dependent on something that could buck us off so easily.
[Image Credit: Brandon Oliver]