Facebook scared a lot of publishers earlier this month when it announced its plans to shake up how often certain sites appear in users’ News Feeds. At a time when hyper-viral posts from Buzzfeed and Upworthy are dominating our streams, Facebook says it wants to make it easier for “high-quality content” to get back on top.
But rather than monitor user activity to figure out what people want to see, Facebook is going in a different, more low-tech direction: It’s asking users directly.
Some Facebook users visiting the site today are being asked to take a survey about the kind of posts they’d like to see in their News Feeds. Of the ten story examples Facebook gave me, four were links to external news sites (NPR, Boston Globe, The Guardian and Wordsmith.org), four were short status updates, one was an Instagram photo, and one was a Buzzfeed link (to a post about cats, naturally).
Facebook says that user input helps it show users “more of what they like,” which is a telling insight into how Facebook defines a “better” News Feed experience.
“We’re trying to align our definition of value with that of our users,” News Feed manager Lars Backstrom told AllThingsD earlier this month.
Facebook’s survey is in many ways a clear response to one of the biggest issues with basing its News Feed changes solely on user activity: Clicks and shares may show Facebook that people are in love with Buzzfeed and Upworthy, but actual user survey responses might point in another direction entirely.
Or perhaps the opposite is closer to the truth: People may say they want more high-quality links in their News Feeds, but the popularity of Upworthy and Buzzfeed say something very different.
That’s not lying, exactly, but it underscores what social media has always been: A way for people to show the world who they want to be perceived as, not precisely who they are. It’s a reality that predates Facebook itself.
This, again, exposes one of the big problems with Facebook’s data-driven approach to everything: While the company is very good at understanding the “what” of user activity (what people click on, what pages they visit), it’s not so good at understanding the “why.” Perhaps it’s looking to turn that around.
VB’s research team is studying mobile user acquisition...
Chime in here, and we’ll share the results