Why the NYT will lose to HuffPo
Tom McGeveran asks an important question, in his analysis of the AOL-HuffPo deal:
What is it about the environment of traditional journalism that makes it so that readers are more likely to interact with the Huffington Post reblog of a New York Times article than they are with the article itself?
The answer to this question, I think, is also a key part of the reason why the NYT paywall is a bad idea.
It’s worth using a specific example here, so let’s take Dave Pell’s suggestion and look at the NYT’s Olbermann scoop last night, and HuffPo’s reblog of it. When Pell first tweeted the comparison, the NYT blog had no comments, while the HuffPo blog had “hundreds of comments/likes.” Now, the NYT post is up to 93 comments, but the HuffPo post is still miles ahead: 2,088 comments, 1,392 likes on Facebook, 340 Facebook shares, 89 tweets, and 52 emails. All of which figures are easily visible in a colorful box at the top of the story.
The NYT, by contrast, keeps such numbers to itself: you can see the number of posted comments, but you can’t see the number of comments which have been submitted and have yet to make it through moderation. (Which is why Pell saw zero comments when he tweeted last night.)
Both the NYT and HufPo stories are blog posts, but there the similarities end. It’s worth just looking at the two, side by side:
McGeveran says the NYT doesn’t look more like HuffPo because “their very existence is justified by their obligation to readers to vouch for the content they produce,” including widgets from Twitter and Facebook and the like, if the content from those widgets appears on the NYT website. And I daresay that the NYT sells that Killington ad at a much higher rate, per thousand pageviews, than HuffPo is getting for its ProFlowers and Lufthansa ads and its Bing tie-in. But the HuffPo page is clearly generating lots of pageviews, and has more ads on the page. (The second ad unit on the NYT page is a house ad for its own iPad app; HuffPo also has a house ad, for its Blackberry app, but runs it separately in a blue bar along the top of the page.) And of course HuffPo doesn’t need to pay for the expensive original reporting of Bill Carter and Brian Stelter.
Still, the difference between the two pages is much starker than it needs to be: the NYT page is like walking into a library, while the HuffPo page is like walking through Times Square. The HuffPo page is full of links to interesting stories elsewhere on the site — about Egypt, or the kid in the Superbowl Darth Vader ad, or the stories my Facebook friends are reading. And there are lot of links to media stories, too; each one has a photo attached.
The NYT page, by contrast, feels like it’s at a site-map dead end. It’s part of the Media Decoder blog, and almost every NYT story linked to on the page is also part of that blog. There are almost no photos; there is almost no color.
Most importantly, the HuffPo page is genuinely, compellingly, interactive — it’s almost impossible to visit it without finding something you want to click on. Like! Comment! Tweet! Go here! Try this! Visit that! There’s site navigation, yes, but that’s just one layer of a very rich and complex page architecture. At the NYT page, by contrast, to get out of the Media Decoder blog you either have to click on a generic navigation button like “Sports,” or else you’ll just leave the page and the site completely.
At this point, it’s worth remembering that the NYT paywall is really, at heart, a navigation fee. The side door is always open: if you get to the NYT website from, say, the HuffPo story, then you’ll be able to read that story no matter how many other stories you’ve read that month. The NYT has said that 80% of its visitors don’t read enough pages per month that they’d have any need to subscribe at all. But it’s pretty obvious why that is: the NYT is making precious little effort to encourage people to want to click around the site and view more pages.
The fact is that readers come to the NYT — or any website — because they want to read its stories. They don’t much care about branded sections, or deciphering the difference between a news story and a blog entry. (The Olbermann story is a blog post, for reasons which even I don’t fully understand.) But the NYT site architecture seems built around the peculiar way that the news is produced inside 620 8th Avenue, rather than around showing the NYT’s readers the exact stories they’re most likely to want to read.
One the paywall goes up, it’s certain that non-subscribers — the vast majority of the NYT readership — will read fewer pages per month than they did before. The NYT’s navigation was never very sticky, as we’ve seen, but from here on in there will actually be a substantial economic incentive not to explore the site, and to save up your precious quota of pageviews for when you need them most.
One of the paradoxes of news media is that most of the time, the more you’re paying to use it, the harder it is to navigate. Sites like HuffPo make navigation effortless, while it can take weeks or months to learn how to properly use a Bloomberg or Westlaw terminal. Once the NYT implements its paywall, it’s locking itself into that broken system: it will be providing an expensive service to a self-selecting rich elite who are willing to put in the time to learn how to use it. Meanwhile, most Americans will happily get their news from friendlier and much more approachable free services like HuffPo.
Rather than learning from or trying to emulate HuffPo’s hugely valuable editorial technology, then, the NYT is sticking its head in the sand and retreating to a defensive stance of trying to make as much money as possible from its core loyal readers. There’s no growth in such a strategy. Indeed, the opposite is true: the NYT is making it both hard and expensive to become a core loyal reader. Meanwhile, the open web will become ever more accessible and social, with friends pointing friends to news in a site-agnostic manner. The NYT is distancing itself from that conversation, standing proud and aloof. It’s a strategy which is doomed to fail.