Are Web metrics corrupting online media?

Are Web metrics corrupting online media?

July 20, 2015

We’re sup­posed to be living in an “attention econ­omy,” right? But standard Web metrics don’t really measure attention. Uniques, page views, monthly aver­age users—none of these take into account how long a visitor spends, whether the visitor is satisfied, or even if the visitor is human and not a bot. For all our talk of Big Data, au­di­ence measure­­ment remains a remarkably crude sci­ence. And yet, because ad­ver­­tisers and in­vestors rely on these metrics, they have a distorting effect on social media and pro­fessional media alike. This has been one source of Twitter’s problems, but it’s even worse with online journalism. If the sites with the most uniques win, it’s easy to game the system by baiting users with value-free content. So cheese­ball aggregators like Viral­Nova—which just sold for something approaching $100 million—get rewarded, while more serious stuff is penalized.

I’ve been fasci­nated by this phenom­enon at least since 2009, when my colleague Daniel Roth pub­lished an excellent story in Wired called “The Answer Factory.” It was about Demand Media, a self-aggrandizing start­up whose stated mission was to pub­lish “what the world wants to know and share”—which turned out to be a cynical euphem­ism for “what we can stuff with ads and trick peo­ple into click­ing on.”

Today, De­mand Media is a cautionary tale. Largely re­li­ant on search results for its traffic, it had its air sup­ply cut off by Goo­gle just a few weeks after a fren­zied IPO. Copy­cats suf­fered the same fate. So enterprising Web pub­lishers switched their dis­tribu­tion strategy from search to social, and sud­den­ly we started get­ting link­bait sites like A+ and Viral­Nova, which describes it­self as host­ing “the web’s most fasci­nating, inspir­ing, and emo­tionally strik­ing stories.” This worked great, at least for them, until Face­book changed its News­Feed algo­rithm (just as Goo­gle before it had changed its search algo­rithm) and traffic took a nose­dive. I suspect it’s no coin­cidence that Viral­Nova now says it’s developing more original content.

What’s behind all this? That’s the question I set out to answer in “The Attention Economy 3.0,” just out in the summer issue of the Milken Institute Review. My con­clusion: The rise of click­bait and linkbait is a direct re­sult of the fail­ure to measure what really mat­ters:

The focus on uniques and click-throughs assumes that value online is where it’s always been in the media business, in ad inventory. But Web pages can mul­tiply like bunny rab­bits—and the more they do, the less valuable each ad oppor­tunity be­comes. What’s needed is an in­jec­tion of scarcity—and “the only unit of scarcity on the web,” as Chart­beat CEO Tony Haile recently told Advertising Age, is time.

Time is what the at­ten­tion economy is all about—the idea that because we live in an era when informa­tion, former­ly a scarce com­­mod­ity, is su­per­abundant, peo­ple’s at­ten­tion, once abun­dant, has be­come scarce. From dia­monds to lead­ership skills, we value what is scarce—so in the atten­tion econ­omy it’s no longer con­tent that’s val­uable, it’s the atten­tion peo­ple give that con­tent. This is a rad­ical no­tion, a com­plete re­ver­sal of the val­ue proposi­tion that pre­vi­ously un­der­lay all media. But it’s been es­sen­tially ig­nored in fa­vor of metrics that treat dig­ital me­dia the same way Niel­sen ratings al­ways treat­ed TV.

Tony Haile and others have made it their mis­sion to cre­ate metrics that re­flect this re­ali­ty. They say this will im­prove the qual­ity of in­for­ma­tion we get on­line. So do Web pub­lish­ers like Me­di­um’s (and Twit­ter’s) Evan Williams and even Upworthy‘s Peter Koechley. I think they’re right, and in “The Atten­tion Economy 3.0” I try to explain why. Please let me know what you think.

The Sea We Swim In

"It’s a zingier version, then, of the post-Aristotelian story-theory books beloved of screenwriters, with a rich range of reference that takes in the novels of Gustave Flaubert as well as the twists of ABC's 'Lost.' But the analysis has a wider salience. . . . It’s critical thinking for an age of pervasive media."

— Steven Poole, The Wall Street Journal

Comments

Sarah Kershaw

- July 22, 2015

Hello,

It's wonderful to get your ideas on this subject as I am fascinated by this subject. That might sound impossibly dry but there it is!

I'm not sure if you have any connection with Fast Company, but back in 2013 I watched a webinar hosted by one of their editors, Chris Dannen.

One figure that stayed with me is that Fast Co Labs managed to increase their time on site metric (a much better metric than "uniques" as per your article) by 42%, by experimenting with the live blog format for news stories. To recap for the purposes of being useful to your readers, a live blog in this instance is as follows:

"On one URL, an article is posted and more content is added repeatedly in either chronological or reverse chronological order. Video and comments are almost always featured on the page to supplement the copy.

They were doing this because they were taking a long view on which metrics might become more valuable, as algorithms evolve. Of course, there are other advantages to the live blog format too, such as A/B headline testing to name just one.

I thought I'd mention it just in case it was helpful. Meanwhile I'm trying to get an update on how this experiment is faring.

More from the Blog