Facebook Takes a Cue from the TV Industry: Poll the People

Jacob Harper |

As much as Facebook Inc. (FB) is the new boss in town, the more they grow the more they’re like the old boss. This has become more apparent as Facebook continues to trust their wall algorithm less, and trust that old standby of media – the poll – more and more.

Think of Facebook like a network TV station and the Facebook News Feed as the station’s programming slate. To be sure, Facebook has done pretty well tailoring each individual’s “channel” to their liking using two elements: the user’s self-selected content producers (i.e. the user’s friends), and those content producer’s “best” material; that is, the stuff that gets a lot of likes and comments. Facebook has little else to do to their user’s channel; the algorithm does the rest, ostensibly maximizing returns by producing a feed that keeps users glued to Facebook for the longest time possible.

But there’s a problem: What if their feed, tailored for that potent mixture of maximum enjoyment and maximum advertising revenue, malfunctions and annoys a user so much they leave the site entirely? Churn is a real problem, one Facebook has begun addressing at first by diversifying and acquiring and producing different “channels.”



But there’s still the issue of the flagship station, and people not being as happy with their feed as they had been. Even if the algorithm had been designed just so to highlight the “right” posts while burying the “wrong” ones, something was wrong. So Facebook has responded like any good old media company: by asking users why they like and don't like things.

When users hide a post on Facebook now, a prompt will ask them what they don’t like about that particular post, or even that content producer. And then it tweaks the feed that much more.

Think of it like a test audience. TV stations have for decades employed the “pilot” system to gauge what audiences like and don’t like. After a showing, they ask the test viewers: what was it about this character that resonated with you? What was it about that storyline you didn’t like? In a similar fashion, by factoring that information in with the reams of data they’ve accumulated as to users tastes (culled from user’s likes, comments, and hides) Facebook can then continue perfecting the algorithm, and minimizing “shows” (i.e. posts) that tend to annoy people enough that they change the channel.

It’s a throwback to the old way of tailoring content, while building on the success Netflix (NFLX) has employed to produce their own original content. Netflix famously gave the green light the show House of Cards after their algorithm determined that a political drama starring Kevin Spacey and directed by David Fincher would do exceptionally well.

As Facebook continues to grow, they must attract more users, and keep them on the site (or one of their subsidiary sites like Instagram) longer. But at the very least, they have to keep them from fleeing entirely. And that starts from not guessing as to what users like or dislike via hiding, but asking them, like they did in the old days, “why do you feel how you do, and how can we change to please you?”

DISCLOSURE: The views and opinions expressed in this article are those of the authors, and do not represent the views of equities.com. Readers should not consider statements made by the author as formal recommendations and should consult their financial advisor before making any investment decisions. To read our full disclosure, please go to: http://www.equities.com/disclaimer

Companies

Symbol Name Price Change % Volume
FB Facebook Inc. 146.49 1.02 0.70 17,770,510 Trade
NFLX Netflix Inc. 152.16 8.33 5.79 20,678,198 Trade
GHIL Green And Hill Indus Inc 0.00 0.00 0.00 0

Comments

Emerging Growth

Cavitation Techs Inc

Cavitation Technologies Inc engineers and designs Nano technology based systems that have potential applications in industrial liquid processing.

Private Markets

Initial State

Initial State is an Internet of Things (IoT) data analytics & data management platform company. We turn sensor and event data into information that matters by making it easy to…

CF0005, LLC dba Mesa-Marshall #1-2

MESA Resources, Inc. ("MESA") is sponsoring the development of the oil & gas well completion project "Mesa-Marshall #1-2". MESA, founded in 1992, is engaged in the acquisition, exploration and development…