Content Discovery: Quality Vs. Quantity
Is it better to have one related story or 150 of them?
As time spent gains ground on page views as the universal metric for engagement, publishers are trying to figure out the best ways to keep readers on their sites. It’s still up for debate whether that’s best done with a clean, uninterrupted experience, or by using a shotgun method and a blast of related content.
Regardless of the approach they’re opting for, optimizing the content discovery process is an overriding theme—Time Inc.’s multiyear, $100-million deal with Outbrain highlights just how important routing readers to the next story has become. Two-by-three, bottom-of-the-story widgets and "Most Popular" rails have been ubiquitous for a while now; full homepage footers and continuous scroll are increasingly finding their way into redesigns.
Popular Science opted for the latter in its recent site relaunch, unveiled last week. Continuous scroll and a single-column, minimalistic black-and-white design get out of the reader’s way.
Cliff Ransom, PopSci’s editor-in-chief, says it was clear their audience was willing to stick around to go deeper. Its old homepage featured a center column with 10 stories. At the bottom, users would be able to navigate to another page, with a new set of 10 stories.
"They were clicking ‘next,’ then ‘next’ again, then ‘next’ again, and again, and again," he says.
Subbing a scroll for a click-and-load process, it’s a simplified, automated way of encouraging that same behavior.
Article pages were built with the same theory in mind, Ransom says. PopSci’s old site gave users more than 20 options on where to go from any given story—popular articles, photo galleries or archives, for example. On the new site, readers are automatically served the next story in that given content vertical as they scroll down the page.
Additional recommendation options are essentially nonexistent in that format, so, along with ads, they’ll be interspersing "splashy editorial promotions" between articles (the feature is currently available on the homepage, but it just hasn’t been integrated into the article pages yet, Ransom says).
PopSci and Slate (pictured) have the same goals, but they’re 180 degrees apart in their approaches. Between a lengthy right rail, three widgets and feeds for each of its nine verticals, Slate’s article pages give readers up to 150 options of where to go next.
The layout is actually a recent update to its year-old redesign. David Stern, Slate’s director of product development, tested a handful options over the last six months in an effort to get readers beyond the story they came for (the original article page template had a shorter rail and just three recommendations from Slate’s homepage at the bottom).
The winner—the 150-link option—went live in early November.
"Anecdotally, we were hearing from users that there wasn’t a lot to click on once they got to the bottom of the page, so we figured more options would increase the chances that you’d find something of interest," Stern says. "It wasn’t a huge bump, but whether you look at page views or time on site, this was a net win."
As Stern says, there wasn’t a huge difference between any of the versions. Testing showed a 2.59-percent increase in clicks on the new regions of the page, and a 0.3-percent increase in overall clicks. He admits it was a smaller gain than they were hoping for, but it’s still a positive alternative.
The downsides—slower load times and an overwhelming feel for the reader—are minimal, he adds. The design is text-centric so there aren’t a lot of large images to delay loading, and the test results show that people aren’t being scared away.
While the volume approach is working for Slate, and evidence pointed PopSci toward a minimalistic design, Ransom and Stern agree that a publisher’s best bet ultimately depends on their readers’ preferences.
"You can imagine different site layouts and different audiences performing differently in whatever context they’re in," Stern says. "Maybe they’re doing the right thing for them, and we’re doing the right thing for us. But we’ve always made these decisions based on data."