Commentary

Death By Magic Beans

Hey, kids, let’s play the Good News/Bad News game! Anybody can do it! All you need is the ability to imagine utopia and dystopia at the same time. Plus, in this case, the ability to give Facebook the benefit of the doubt.

Now I know that last thing might be a bit of a deal breaker. Keeping an open mind about Facebook is like keeping an open mind about chlamydia. But the stakes are very high here -- the highest, really -- so please play along.

At issue is the negotiations with BuzzFeed, National Geographic and The New York Times to host those publishers’ content on Facebook itself. That is: not to link to the source, but to have the content actually reside on Facebook. The originating media company produces the content, but Facebook distributes it and sells the ads against it.

Permit me to tell you my initial reaction when I read about this possibility in the Times one month ago, and I quote: “Holy fucking shit!”

advertisement

advertisement

Why such a reaction? Well, for starters, I am not just a student of media monetization, and a very interested party in media monetization livelihoodwise, I also teach media monetization at a famous-name Ivy League University. This semester, we’ve looked deeply at micropayments, affiliate marketing, data, crowdsourcing, subscriptions, native advertising, noblesse oblige -- in all, 11 revenue streams or business models underwriting everything from the best journalism to the emptiest of content calories. 

Tragically, none of them alone -- nor all of them put together -- seem capable of sustaining more than a small fraction of the media we expect and deserve. You may have read about this somewhere. I call it the Chaos Scenario.

But then I saw the March 23 story and understood instantly: this could be the magic beans everybody has been seeking, the way to achieve vast scale at relatively high CPMs without running headlong into the law of supply and demand. With 1.8 billion users and the world’s biggest dataset, Facebook could deliver both vast audience and valuable narrow targeting beyond even the Times’ mighty reach.

Let's say Facebook treats partner content like iTunes treats songs; it takes 30% of ad sales and passes 70% along to the content owner. This arithmetic doesn't require much imagination: 70% of twice your ad revenue, with no increase in the marginal cost of the goods, is far better than 100% of your existing ad revenue. And 70% of, say, quintuple your ad revenue starts to look like the good old days. This gets us into William Randolph Hearst territory. This gets us into Caligula territory.

In other words: salvation. Woo hoo.

But this is the Good News/Bad News game, and we aren't finished. Now comes the bad news. And I hope I have room.

• Far more than now, the Times and all other Facebook partners would be incented to produce the most shareworthy content for Facebook. Goodbye, Sierra Leone and environmental coverage. Hello, Grumpy cat.

• Facebook would have to be trusted, once becoming the center of the Times distribution world, not to change the rules midstream.  Such as, hypothetically, by tweaking the EdgeRank algorithm to influence which articles get into the newsfeeds of whom -- either to squeeze its partners for a bigger rev share or to deemphasize “problematic” content. Of course, they've never pulled such highhanded stunts -- except for, you know, constantly from the beginning.

• The very fragmentation that has destroyed the media economy has, 'til now, performed a great service. From the 1970s on, one of the great threats to the marketplace of ideas and to democracy itself was deemed to be media concentration. When Gannett, Disney, Time Warner, The Tribune Co. and News Corp. owned such a huge percentage of media properties -- scholars worried -- where would we get diversity of opinion, dissident voices, fresh sources, truly independent journalism? Well, the Internet obliterated those worries.

We have nothing if not a bustling marketplace of ideas. Unless. Unless the distribution of news and commentary is placed predominantly into the hands of Facebook and Google, in which case the worst nightmares of scholars Ben Bagdikian and Robert McChesney would seem like sex dreams.

Or, to revert to the earlier metaphor, imagine the magic beans taking root and producing a soaring beanstalk reaching to the heavens -- where an angry giant or two terrorize us into abject submission. Whereupon our illusory salvation quickly becomes our ruin.

12 comments about "Death By Magic Beans".
Check to receive email when comments are posted.
  1. Martin Focazio from EPAM Systems, April 27, 2015 at 9:24 a.m.

    Facebook produces confirmation bias at a global scale. 

    You think we're a polarized society now? Wait until all media - not just ads - reinforces only what you think is true because you want it to be true. 

    http://en.wikipedia.org/wiki/Confirmation_bias

  2. Kevin Lee from Didit, April 27, 2015 at 11:17 a.m.

    I can't even imagine the fear rippling through the executive ranks at publishers not given this opportunity. Because even if it has a chance of being a disaster the upside mathematics seem compelling at first. However, in the end all advertising lives or dies based on its ability to drive some combination of brand lift or DR lift.  The econometric models get built and the split tests get run.  So, channel of distribution is moot if the consumers don't notice and engage in the advertising. 

  3. Leonard Zachary from T___n__, April 27, 2015 at 11:21 a.m.

    In theory it sounds like it could work. The Elephant in the room called "Audience Fragmentation" could make it a "Lose Control" event and then NYT will not have a paddle to get back. 

  4. Jeff Loman from MultiAd Kwikee, April 27, 2015 at 11:26 a.m.

    "......famous-name Ivy League University" is redundant.

  5. Jonathan Hutter from Northern Light Health, April 27, 2015 at 11:26 a.m.

    I'm skeptical of this arrangement because of your note about the content streams that are deemed "shareworthy" -- junk posts. We get enough of that.

    I'm also skeptical because Facebook's vision (same as Google's and others) still remains, World Domination. Let's control as much information to you and use as much information about you as possible.

  6. Douglas Ferguson from College of Charleston, April 27, 2015 at 11:29 a.m.

    So much fear to be generated, I guess McChesney and Bagdikian couldn't provide enough, so you are lending a hand. I am confident that the mass audience will not forget Sierra Leone. And if not, Facebook is likely already planning Facebook for Serious People, which they can market to elitist universities trolling for rich parents.  I know they have plans for Facebook for Business because the LinkedIn model is too tempting to steal.

  7. Rich Forester from Blayney Media Group, April 27, 2015 at 12:07 p.m.

    Certain societal segments and individuals are already being terrorized into "abject submission" today, aided and abetted by the soft fascisim of social media that is turning harder and harder everyday. Be very careful if you own a pizza restaurant in Indiana! That the NY Times and other news providers want to partner with Facebook, Google, et-al is, unfortunately, a dog-bites-man moment. Often the first ones to fan the flames of intolerance, that they would look to partner with the engines that help support their current agendas is as shocking as discovering gambling in Casablanca. "All the news that's fit to print" will certainly take on a new meaning!  What happens the next time some kid takes their own life due to bullying through social media? Will the story be on the front page of the Times - or below the fold, in section B, on page 24? Or does it make the paper at all? While at a media event a few years ago at MIT, one of their key speakers spoke glowingly about a future where 7 billion people would all be connected through social media. To me, that's a real nightmare.

  8. Brendan Howley from twinfish, April 27, 2015 at 12:31 p.m.

    I'm a former CBC investigative data journalist turned media designer/technologist for 'not only for profits' here in Canada and in the US. 

    My 0.02: McChesney (whom I've interviewed at length) is right about downside and Garfield is right in parsing the upside pieces.

    But here's the thing: mobile is going to determine the winner here, not the platform. Sierra Leone is not just a news issue, it's a context issue...and why you'd trust Facebook with context (esp since that's tied to ads: so's Google's entire biz model) is beyond me. It's like expecting iTunes to give you everything you want—it doesn't. Walled garden. 

    Shouty caps: both Google and Facebook are (ultimately: out of their own founders mouths, on the QT) ARTIFICIAL INTELLIGENCE companies—they're not media companies. They want to be the next grid, but not for energy, for data insight and behavioural data. 

    What will keep them honest is the sheer fragmentation of news across the personal UX ecosystem. I get longform from Medium and Foreign Policy and quartz; the Economist/FT get me hard financial news via my app...you get the idea. Newsrooms aren't dead, the Guardian experiment still loses money but offers serious stuff that digitizes beautifully—and we may be looking at journalism as a public utility, rather than for-profit, long-term, for high volume news.

    Back to mobile as litmus test.

    Google hasn't designed a decent UX for mobile save maps. I don't think Facebook Messenger is going to induce more news consumption—but an app UX that provides a really absorbing news experience will. Personally, I think SnapChat is far closer to a news solution (with Reddit a close second) than Facebook on mobile.

    Nonetheless Garfield's broached a vital point of debate for newsniks like myself and media owners trying to monetize journalism: cui bono? If it's the news consumer, then we have something. If it's Facebook, then I have my doubts we'll have actually brought more money to the production of intelligent, context-rich journalism. 

    In which case, we're going to see a lot more boutique investigative and news shops, funded by private subscription, with small, specialized (almost longtail) audiences but audiences who're highly participatory. VICE is this model on steroids, even if it's relative news value is debatable, as the NYT's late great David Carr made well clear. 

    Great post. Keen to see how this plays out...one wonders how much of news consumption is cultural, rather than demographic. For more on that, see this deep dive from Daniel Kahan and team (2012) http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2193133&rec=1&srcabs=2216469&alg=1&pos=2

  9. David Cutler from EatMedia, April 27, 2015 at 12:46 p.m.

    Classic Garfield... and if you do damned if you don't... But mostly if you do.

  10. Ava Seave from Quantum Media, April 27, 2015 at 3:50 p.m.

    See this New Yorker cartoon about Magic Beans. "These are magic beans... their value comes from growth and scale, not revenue." 

    http://www.condenaststore.com/-sp/These-are-magic-beans-my-boy-Their-value-comes-from-growth-and-scale-n-New-Yorker-Cartoon-Prints_i13192519_.htm

  11. David Vawter from Doe-Anderson, April 27, 2015 at 5:20 p.m.

    "Wait until all media - not just ads - reinforces only what you think is true because you want it to be true."

    Huh?  If that were true how come people profess to dislike advertising so much?  Besides, at least when I see a Windex commercial I know exactly what they're selling and who is behind the placement.  If only the same were true of Fox, msnbc etc.



  12. Paula Lynn from Who Else Unlimited, May 2, 2015 at 6:04 p.m.

    Becoming quite a topic for MediaPost the past few days. The Echos of Cassandra are just beginning. If value comes from growth and scale, not revenue, then the fairy tale is built on ....nothing. What happens when the fantasy leaves in a puff of hot air and the 1% turns into the .0001% who are the only ones who can afford to absorb it all and control it all ?

Next story loading loading..