Trust Me, I'm Lying: Confessions of a Media Manipulator

by Ryan Holiday

  * This one is my favorite, because the thing always happens to be not only not unprecedented but hilariously pedestrian.

  X

  TACTIC #7

  KILL ’EM WITH PAGEVIEW KINDNESS

  THE BREAKTHROUGH FOR BLOGGING AS A BUSINESS was the ability to track what gets read and what doesn’t. From Gawker to The Guardian, sites of all sizes are open about their dependence on pageview statistics for editorial decisions.

  Editors and analysts know what spreads, what draws traffic, and what doesn’t, and they direct their employees accordingly. The Wall Street Journal uses traffic data to decide which articles will be displayed on its homepage and for how long. Low-tracking articles are removed; heat-seeking articles get moved up. A self-proclaimed web-first paper like the Christian Science Monitor scours Google Trends for story ideas that help the paper “ride the Google wave.” Places like Yahoo! and Demand Media commission their stories in real time based on search data. Other sites take topics trending on Twitter and Techmeme and scurry to get a post up in order to be included in the list of articles for a particular event. Even tiny one-person blogs eagerly check their stat counters for the first sign of a spike.

  Bloggers publish constantly in order to hit their pageview goals or quotas, so when you can give them something that gets them even one view closer to that goal, you’re serving their interests while serving yours. To ignore these numbers in an era of pageview journalism is business suicide for bloggers and media manipulators. And anything that pervasive presents opportunities for abuse.

  I see it like this: The Top 10 “Most Read” or “Most Popular” section that now exists on most large websites is a compass for the editors and publishers. Mess with the magnet inside the compass and watch as its owner goes wildly off track.

  As economists love to say: incentives matter. What makes the Most Popular or Most E-mailed leaderboard on Salon.com or the New York Times is a clear directive that tells writers what kinds of stories to head toward. It doesn’t matter that the stories suck or if they have nothing to do with the publisher’s mission. This is about getting pageviews—by whatever means.

  THE DISTURBING SCIENCE

  Yellow papers had their own circulation dragons; instead of celebrity slideshows, these papers had staples like hating black people, preposterous Wall Street conspiracies, and gruesome rape and murder stories. But while in the past decisions were guided by an editor’s intuitive sense of what would pander to their audience, today it is a science.

  Sites employ full-time data analysts to ensure that the absolute worst is brought out of the audience. Gawker displays its stats on a big screen in the middle of their newsroom. The public can look at it too at Gawker.com/stats. Millions of visitors and millions of dollars are to be had from content and traffic analysis. It just happens that these statistics become the handles by which manipulators can pick up and hijack the news.

  It’s too transparent and simple for that not to be the case. For some blog empires, the content-creation process is now a pageview-centric checklist that asks writers to think of everything except “Is what I am making any good?” AOL is one of these organizations, as it emphatically (and embarrassingly) outlined in a memo titled “The AOL Way.” If writers and editors want to post something on the AOL platform they must ask themselves:

  How many pageviews will this content generate? Is this story SEO-winning for in-demand terms? How can we modify it to include more terms? Can we bring in contributors with their own followers? What CPM will this content earn? How much will this content cost to produce? How long will it take to produce?1

  And other such stupid questions.

  Even the famed New Yorker writer Susan Orlean has admitted her gravitational pull toward the stories on the Most Popular lists, as a reader and as writer. “Why, I wonder, should the popularity of a news story matter to me?” she writes.

  Does it mean it’s a good story or just a seductive one? Isn’t my purpose on this earth, at least professionally, precisely to read the most unpopular stories? Shouldn’t I ignore this list? Shouldn’t I roam through the news unconcerned and maybe even uninformed of how many other people read this same news and “voted” for it?2

  But in the end these guilty pangs cannot win out. Amid the clutter and chaos of a busy site, the lists pop. The headlines scream out to be clicked. Those articles seem more interesting than everything else. Plus, hey, they appear to be vetted by the rest of the world. That can occasionally be a good thing, as Orlean points out, but is it worth it?

  Sometimes they contain a nice surprise, a story I might not have noticed otherwise. Sometimes they simply confirm the obvious, the story you know is in the air and on everybody’s mind. Never do they include a story that is quiet and ordinary but wonderful to read. [emphasis mine]

  That great insight is often buried in material that seems quiet and ordinary does not matter to blogging. That wouldn’t get clicks.

  I’m fond of a line by Nicolas Chamfort, a French writer, who believed that popular public opinion was the absolute worst kind of opinion. “One can be certain,” he said, “that every generally held idea, every received notion, will be idiocy because it has been able to appeal to the majority.” To a marketer, it’s just as well, because idiocy is easier to create than anything else.

  THEIR METRICS, YOUR ADVANTAGE

  What gets measured gets managed, or so the saying goes. So what do publishers measure? Out of everything that can possibly be measured, blogs have picked a handful of the most straightforward and cost-effective metrics to rely on (wonderfulness is not one of them). They choose to measure only what can be clearly communicated to their writers as goals. Like officers in Vietnam ordered to report body counts back to Washington as indicators of success or failure, these ill-conceived metrics—based on simplicity more than anything else—make bloggers do awful things.

  To understand bloggers, rephrase the saying as: “Simplistic measurements matter.” Like, did a shitload of people see it? Must be good. Was there a raging comments section going? Awesome! Did the story get picked up on Gawker? It made the Drudge Report? Yes! In practice, this is all blogs really have time to look for, and it’s easy to give it to them.

  I exploit these pseudo-metrics all the time. If other blogs have covered something, competitors rush to copy them, because they assume there is traffic in it. As a result, getting coverage on one site can simply be a matter of sending those links to an unoriginal blogger. That those links were scored under false pretenses hardly matters. How could anyone tell? Showing that a story you want written is connected to a popular or search engine–friendly topic (preferably one the site already has posts about) does the same thing. However tenuous the connection, it satisfies the pageview impulse and gives the blogger excuse to send readers to their stories. You’ve done something that gets them paid.

  Remember, some bloggers have to churn out as many as a dozen posts a day. That’s not because twelve is some lucky number but because they need to meet serious pageview goals for the site. Not every story is intended to be a home run—a collection of singles, doubles, and triples adds up too. Pageview journalism is about scale. Sites have to publish multiple stories every few minutes to make a profit, and why shouldn’t your story be one of them?

  Once your story has gotten coverage, one of the best ways to turn yourself into a favorite and regular subject is to make it clear your story is a reliable traffic draw. If you’re a brand, then post the story to your company Twitter and Facebook accounts and put it on your website. This inflates the stats in your favor and encourages more coverage down the road. There are also services that allow you to “buy traffic,” sending thousands of visitors to a specific page. At the penny-per-click rates of StumbleUpon and Outbrain, one hundred dollars means a rush of one thousand people or more—illusory confirmations to the blogger that you are newsworthy. The stat counters on these sites make no distinctions between fake and real views, nor does anyone care enough to dig deep into the sources of
traffic. The lure of the indirect bribe is all that matters.

  But be careful: This beast can bite you back if it feels like it. Once sites see there is traffic in something, they do not stop—often falling to new lows in the process. Companies enjoy the spotlight at first, until the good news runs out and the blog begins to rely on increasingly spurious sources to keep the high-traffic topic on their pages. What begins as positive press often ends in the fabrication of scandals or utter bullshit. As Brandon Mendelson wrote for Forbes, the lure of pageviews takes blogs to places they otherwise never should have gone:

  A couple of years ago, I quit blogging for Mashable after they had posted the suicide note to the guy who flew a helicopter into a government building in Texas. Pete’s [the publisher] response to me quitting over the suicide note was, pretty much, “Other blogs were doing it.” He never explained why a Web / Tech / Social Media guide would post a crazy person’s suicide note.

  “Who wants to say ‘I did it for the page views’ out loud?”3

  The answer to that question is “almost every blogger.”

  Why do you think the Huffington Post once ran a front-page story about what time the Super Bowl would start? The query was a popular one on game day, and the post generated incredible amounts of traffic. It may have been a pointless story for a political and news blog like the Huffington Post to write, but the algorithm justified it—along with the rest of their “the world is round” stories and well-timed celebrity slideshows.

  This content is attractive to blogs because the traffic it does is both measurable and predictable. Like a fish lure, it is not difficult to mimic the appearance of these kinds of stories and for unthinking writers to fall for it. They are looking to eat. They know what key words are lucrative, what topics get links, and what type of writing gets comments, and they’ll bite without asking themselves whether the version of events you’ve presented is just a barbed trick.

  Metrics and measurements are a comfort to publishers. It takes the uncertainty out of their business. What can’t be measured—or requires true editorial judgment—is scary and requires financial risk.

  CAN’T STAND THE SILENCE

  “I posted something but nobody responded. What does it mean?” It’s a question you’ve probably asked yourself after nobody liked the Facebook status with your big news, or no friends commented on your new Facebook photo album. Maybe you thought that tweet you wrote was hilarious, and you’re not sure why it wasn’t retweeted—not even once. This innocent little question is just about hurt feelings for you, but for pageview-hungry publishers, it’s what keeps them up at night.

  Early Usenet users called this Warnock’s Dilemma, after its originator, Bryan Warnock. The dilemma began with mailing lists but now applies to message boards (why is no one responding to the thread), blogs (why hasn’t anyone commented?), and websites (why isn’t this generating any chatter?). The answer to any of these questions could just as easily be satisfaction as apathy, and publishers want to know which it is.

  This dilemma was actually predicted by Orson Scott Card in the 1985 book Ender’s Game. Peter Wiggin creates the online persona of a demagogue named Locke and began to test the waters by posting deliberately inflammatory comments. Why write this way? his sister asked. Peter replied: “We can’t hear how our style of writing is working unless we get responses—and if we’re bland, no one will answer.”

  Card understood that it is incredibly difficult to interpret silence in a constructive way. Warnock’s Dilemma, for its part, poses several interpretations:

  1. The post is correct, well-written information that needs no follow-up commentary. There’s nothing more to say except, “Yeah, what he said.”

  2. The post is complete and utter nonsense, and no one wants to waste the energy or bandwidth to even point this out.

  3. No one read the post, for whatever reason.

  4. No one understood the post but won’t ask for clarification, for whatever reason.

  5. No one cares about the post, for whatever reason.4

  If you’re a publisher, this checklist causes more headaches than it cures. It’s all bad. Possibility number one is unprofitable: We know that practical utility doesn’t spread, and posts that don’t generate follow-up commentary are dead in the link economy. Possibility number two is embarrassing and damaging to the brand. Possibility number three is bad for obvious reasons. Possibility number four means the post was probably too ambitious, too academic, and too certain for anyone to risk questions. Possibility number five means somebody chose the wrong topic.

  Whatever the cause, the silence all means the same thing: no comments, no links, no traffic, no money. It lands the publisher firmly in a territory labeled “utterly unprofitable.” Jonah Peretti, for his part, has his bloggers at BuzzFeed track their failures closely. If news doesn’t go viral or get feedback, then the news needs to be changed. If news does go viral, it means the story was a success—whether or not it was accurate, in good taste, or done well.

  That is where the opportunity lies: Blogs are so afraid of silence that the flimsiest of evidence can confirm they’re on the right track. You can provide this by leaving fake comments to articles about you or your company from blocked IP addresses—good and bad to make it clear that there is a hot debate. Send fake e-mails to the reporter, positive and negative. This rare kind of feedback cements the impression that you or your company make for high-valence material, and the blog should be covering you. Like Peter Wiggin, publishers don’t care what they say as long as it isn’t bland or ignored. But by avoiding the bad kind of silence prompted by poor content, they avoid the good kind that results from the type of writing that makes people think but not say, “Yeah, what he said. I’m glad I read this article.”

  Professional bloggers understand this dilemma far better than the casual or amateur one, according to an analysis done by Nate Silver of unpaid versus paid articles on the Huffington Post. Over a three-day period, 143 political posts by amateurs received 6,084 comments, or an average of just 43 comments per article (meaning that many got zero). Over that same period, Huffington Post published 161 paid political articles (bought from other sites, written by staff writers, or other copyrighted content) that accumulated more than 133,000 comments combined. That amounts to more than 800 per article, or twenty times what the unpaid bloggers were able to accomplish.5

  According to Huffington Post’s pageview strategy, the paid articles are indisputably better, because they generated more comments and traffic (like a 2009 article about the Iranian protests that got 96,281 comments). In a sane system, a political article that generated thousands of comments would be an indicator that something went wrong. It means the conversation descended into an unproductive debate about abortion or immigration, or devolved into mere complaining. But in the broken world of the web, it is the mark of a professional.

  A blog like the Huffington Post is not going to pay for something that is met with silence, even the good kind. They’re certainly not going to promote it or display it on the front page, since it would reduce the opportunity to generate pageviews. The Huffington Post does not wish to be the definitive account of a story or inform people—since the reaction to that is simple satisfaction. Blogs deliberately do not want to help.

  You’re basically asking for favors if you try to get blogs to cover something that isn’t going to drive pageviews and isn’t going to garner clear responses. Blogs are not in the business of doing favors—even if all you’re asking is for them to print the truth. Trust me, I have tried. I have shown them factories of workers whose jobs are at risk because of inaccurate online coverage. I have begged them to be fair for these poor people’s sake. If that didn’t make a difference, nothing will.

  BREAKING THE NEWS

  I don’t know if blogs enjoy being tricked. All I know is that they don’t care enough to put a stop to it. The response to sketchy anonymous tips, in my experience, is Thanks, a lot more often than Prove it.

  Nobody i
s fooling anyone. That’s not the game—because sites don’t have any interest in what they post, as long as it delivers pageviews. Samuel Axon, formerly an editor at Mashable and Engadget, complained that the rules by which blogs get “traffic, high impressions, and strong ad revenues betray journalists and the people who need them at every turn.” This is only partially true. They betray the ethical journalists and earnest readers. As far as bloggers and publishers looking to get rich or manipulators eager to influence the news are concerned, the system is just fine.

  Pageview journalism puffs blogs up and fattens them on a steady diet of guaranteed traffic pullers of a mediocre variety that require little effort to produce. It pulls writers and publishers to the extremes, and only to the extremes—the shocking and the already known. Practicing pageview journalism means that a publisher never has to worry about seeing “(0) Comments” at the bottom of a post. With tight deadlines and tight margins, any understanding of the audience is helpful guidance. For marketers, this is refreshingly predictable.

  It just happens that this metric-driven understanding breaks the news. The cynicism is self-fulfilling and self-defeating; as the quip famously attributed to Henry Ford points out, if he’d listened to what his customers “said” they wanted, all “we’d have ended up with was a faster horse.”

  Pageview journalism treats people by what they appear to want—from data that is unrepresentative to say the least—and gives them this and only this until they have forgotten that there could be anything else. It takes the audience at their worst and makes them worse. And then, when criticized, publishers throw up their hands as if to say, “We wish people liked better stuff too,” as if they had nothing to do with it.

  Well, they do.