Lateral Movement at the NY Times
August 15th, 2005
The NY Times doesn’t want you to leave their site.
Their advertisers only pay them if you see an ad. The longer you stay on the site, the more money they make from your visit. They definitely don’t want you to leave.
But the Times (and many others) are running into a problem. Because of search engines and content aggregators, people are visiting fewer pages. The drive to content has become more surgical and visitors are employing a “get in and get out, nobody gets hurt” approach.
For the NY Times, that means that the people coming from the RSS feed or email alerts are probably only reading those articles they click on and not exploring the rest of the site. Bad news for advertising revenues. What’s a newspaper to do?
Well, in a valiant attempt to get users to explore the site, the designers at the NY Times added a simple link to each article:
Problem is that the link doesn’t really give off much scent. While it directs the reader to the next article in the series, there are no trigger words to help the reader decide if it’s worth clicking. Therefore, it’s unlikely there would be much movement.
We have a name for the navigation that moving from one content page to another: lateral links. The NY Times is trying to encourage lateral movement, but it’s unlikely these links will really do much. After all, what is the reader going to see in article #5?
The NY Times isn’t the only one with the problem. A recent email campaign from Universal Studios encouraged readers to become contestants on their theme-park version of Fear Factor:
Clicking on the links brought readers to a page describing the Fear Factor attraction. At the bottom of the page is this navigation:
But what does “Previous” attraction mean? After all, the reader just came from an email — not from another page on the site. And, does “more attractions” mean a list of related attractions? (In fact, it brings you to the page for Animal Planet Live. How likely is a person interested in Fear Factor also going to be interested in Animal Planet? Hard to say, but I bet it’s not even close to 100%.)
CNN makes it’s lateral movement more trigger word friendly, by listing the names of the articles. For example, at the bottom of an article about Demi Moore declaring her love for Ashton Kutcher (and why wouldn’t she? He’s such a dream) we find these links:
Click to see CNN’s lateral navigation links.
CNN includes both an entertainment section and a top stories section, hoping to increase the probabilities that one of those will be interesting enough to keep the reader’s attention.
Of course, the grand daddy of cool lateral movement links is Amazon. Amazon is always coming up with ways to keep you moving from one content page to the next. The most famous of which is the “Customers who bought this have also bought…” links that are on practically every page (unless no-one has ever bought the item). For example, here’s the links from the Dude, Where’s My Car? DVD:
Click to see Amazon’s lateral links.
But, look at any Amazon content page and you’ll see a bundle of lateral links, such as the recently viewed links:
Lateral movement is an important design element, but I’m thinking the NY Times has a way to go before they get something that really works for them. After all, how hard would it be for them to create a story recommendation engine?
“Readers who’ve read this story have also read these articles…”
Update: More thoughts on lateral links here.
August 15th, 2005 at 2:11 pm
This is excellent, Jared! And this seems like an excellent example of situations where the designer’s “mental model” of how the site will be used doesn’t correspond well with the user’s “mental model.” The user doesn’t have a “next” and they definitely do not have a “previous.” The labelling makes no acknowledgment of the user’s mental model, and since they don’t seem to understand the different mental models, they aren’t doing any teaching; they aren’t helping the user to participate in the designer’s (and the system’s) mental model. (Sure, this isn’t the best strategy, but it’s certainly better than what they’ve got so far).
“Earlier articles in today’s Business Section”
By setting up the relationship as time-based, and by indicating the location, they can provide some navigational cues to help the user build that model.
I prefer your suggestions above, of course.