Why Facebook’s EdgeRank is bad for user experience and businesses

In Brian Solis’ recent blog post “What Your Business Needs to Know About Facebook’s EdgeRank,” he makes a case for businesses to understand the EdgeRank algorithm before making judgments about the merits of Facebook’s filtering mechanism.

EdgeRank’s algorithm is described as follows:

Affinity (repeated interaction) x Weight (time spent interacting) x Decay (content becomes less valuable throughout time) = EdgeRank

Debating the merits of EdgeRank is somewhat futile:  the “hacker way” is not democratic.  However, understanding why EdgeRank is such a raw deal for users and for businesses is important, particularly as businesses consider additional budget to make up for lost reach.

Social media consumption far surpasses social conversation

Consider these facts:

  • NPD DisplaySearch forecasts tablets to outsell laptops in 2013.
  • Half of user time spent on Facebook is accessed via mobile devices.
  • Most Facebook users consume more than they produce.

I was listening to NPR the other day and they were talking about the astronomical growth in the tablet market.  The big takeaway: tablets are devices designed primarily for consumption rather than creation.  I suspect that the proportion of consumption-to-creation on tablets increases with user age (recall that the average age of a Facebook user is around 40).

When people say “Facebook can’t replace real conversation,” they’re absolutely right.  User behavior significantly favors the media aspect over the social aspect of Facebook (and any other social network).  The rise of tablet and mobile devices increases consumption without interaction and in the context of EdgeRank this is significant.

On a tablet or mobile device, it’s more difficult to “engage” a post.  On the mobile Facebook app (or the desktop to an extent), a user can view content of various qualities without sending any social signals to Facebook.  Facebook “superusers” should be able to exert disproportionate influence over the content that is presented in their friend’s newsfeeds.  Users likely have preferences that aren’t included in PageRank.

Ignore for a minute that Facebook is now deciding what content people consume based upon a universal criteria.  The argument for “Social Media Optimization” is weakened further when you consider that the average user has 130 friends, gaining 7 per month.  Google or Bing provide order to hundreds-of-thousands if not millions of sources of loosely-relevant information for any given search: that’s valuable.  Assuming that each friend posts two times per day (most don’t): a typical user could probably go through all of their friend’s content daily in a matter of minutes.  The amount of content that any user can consume probably doesn’t warrant a filter; social and search are quite different in that regards.

User experience suffers by having to visit each friend’s page directly to see what they’re publishing or missing out on it altogether.  Predicating a spam filter on social signals ignores how users consume content, and EdgeRank will get more irrelevant as mobile and tablet use increases.

The new long tail

In Chris Anderson’s book The Long Tail, he describes how e-commerce has made a near-infinite complement of digital and physical products accessible to anyone.  Wikipedia describes a “long tail” like this:

“In “long-tailed” distributions a high-frequency or high-amplitude population is followed by a low-frequency or low-amplitude population which gradually “tails off” asymptotically. The events at the far end of the tail have a very low probability of occurrence.”

Ignore user experience for a moment (if you pretend you’re Facebook that might be easier).  Facebook Pages meander on a new long tail – the long tail of Facebook’s EdgeRank. Given the EdgeRank algorithm, given that brands get less than 1% engagement on their posts, and given that brands cannot initiate a direct conversation with their fans: this shouldn’t be a surprise.


The Long Tail
The Long Tail

The high-amplitude population on the distribution curve are friends.  You interact with them more than with brands (one hopes), you spend more time with their content (one hopes) and there is always fresher content in the queue – so brands lose the PageRank trifecta.  Without a means to connect directly with fans, brand content is disadvantaged.  Given an increased proclivity for content consumption without social signals, brands are further disadvantaged with PageRank.

In Solis’ post he says brand posts were down 40% after the PageRank algorithm.  If it was only 40%, that’s a good news story for brands.  But for Facebook to imply that the reach can be increased by 67% through better content is disingenuous.

There’s no free lunch for businesses on Facebook anymore: companies need to determine whether they’ll pay-to-play on Facebook.

Jim Dougherty

Jim Dougherty

Writer and chief of miscellany at leaderswest.com
I'm the guy that wrote the article you just read. Sorry for the typos.
  • http://www.arielmarketinggroup.com/ Amy McCloskey Tobin

    I know there are many ‘experts’ who tell me I’m wrong,but I think that Facebook is making it so difficult for small businesses to reach their followers that they have to minimize the importance of Facebook as a marketing/communication channel. I know, FB needs revenue, and that will come from Big Brands… but as my feed becomes more and more cluttered with Sponsored Stories, I start to ignore them just like I do PPC ads on Google. I’m with you – FB Edgerank is a raw deal for ‘us.’

    • jimdougherty

      Thanks Amy – you’ve been super generous with your time and comments and it is always appreciated. I’m there with you on business, and I find it a very flimsy argument to say that these changes are for the benefit of users. Social signals with Facebook (dating to Poke 1.0 and Likes) have always been opaque, and I don’t think there is evidence that people’s comments are sufficient signal to show preference for one piece of content over another. To your point about your news feed being overrun with ads, the WSJ ran a piece about the uptick in Facebook’s stock stating explicitly that Facebook’s revenue does better as the percentage of ads to organic content increases. Interesting that they can show so much savvy for paid advertising yet have no financial interest in filtering results, huh?

  • http://blessayfromamerica.blogspot.com/ Guy Bailey

    You are absolutely right Jim. I work as a social media specialist for a University in GA and our engagement rates dropped off a cliff with the new rank. We have steadily built it back up and but we will never pay for a promoted post or tweet.

    It’s extortion – “nice post you have here, be a shame if nobody saw it…”

    • jimdougherty

      Thanks Guy for your great comment. I struggle with calling it extortion, though. I think it’s indicative of the pro forma business strategy for most social networks. Great idea generates investor dollars which generates scalable platform which generates a monetization effort. Facebook has been under extraordinary pressure since their IPO and Twitter will be in the same boat in the next year when they go public, so I’m hopeful someone comes up with a third way so that users aren’t put in a position where they feel extorted or that the brand promises these platforms make are broken.