Slate’s Long Tail

Does Chris Anderson’s theory apply to online magazines?

Listen to an interview with the author here, or sign up for Slate’s free daily podcast on iTunes.

When Wired editor-in-chief Chris Anderson published an article called “TheLong Tail” two years ago, he was hailed for explaining how the Internet changed the way culture purveyors do business. When he published a book with the same name this month, Tim Wu took him to task for overreaching—unleashing his nifty theory on everything from eBay to al-Qaida, whether it applies or not. But there’s at least one indispensable American brand that has thus far escaped the long arm of the Long Tail: Slate. Can Anderson’s theory shed light on the economics of an online magazine?

Slate seems like a classic Long Tail candidate. Anderson’s basic idea is this: The Internet allows online businesses to maintain vast inventories, because it’s cheaper to store bits on a server than to keep, say, CDs on retail shelves. That means consumers now have access to products that aren’t popular enough to warrant space in bricks-and-mortar stores. What’s interesting is that they’re buying these products in droves. Anderson reports that 25 percent of Amazon’s total sales come from titles not stocked in retail outlets, which means the company is less dependent than its offline competition on best-selling hits. 

At Slate, our inventory is our articles. We publish 20 or so stories every weekday, but we also have a backlog of about 33,000 pieces in our archives. Because those stories are freely available to our readers, a chunk of our traffic each day comes not from our “hits”—current pieces that are promoted on our home page, which typically draw tens of thousands of readers—but from older pieces with narrower appeal. The economist’s question is: How big is that chunk? And the editor’s question is: Which of our hoary old chestnuts are you reading?

The answer to the second question is easy: You’re reading the one about Girls Gone Wild. In March of 2004, Slate published a series of dispatches by Ariel Levy about the making of a Girls Gone Wild video. Levy, who later used some of the material in Female Chauvinist Pigs, her excellent book on raunch culture, wrote about the politics of exhibitionism—and about girls getting spanked on South Beach. Levy’s piece got at least 1,700 page views a day last week—racking up more than 13,500 altogether—even though there was no link to it anywhere on Slate’s home page.

Why did the piece pull in such consistent numbers? Google. When readers type “Girls Gone Wild” into Google’s search box—seeking intellectual succor, no doubt—Levy’s piece is the fourth hit. The story here, of course, is not that Levy’s dispatches got 2,354 hits last Tuesday; Slate as a whole pulled in about 1.9 million hits that day. The story is that that traffic was free—we paid for the piece years ago, and we didn’t expend any additional man-hours last week assigning, editing, or producing it. That means Levy’s dispatches provided 2,354 chances for our advertisers to reach our readers—and pay us for the privilege of doing so—without costing us a thing.

There are plenty of other Google hits in Slate’s archives. Paul Boutin’s 2004 piece about how to steal Wi-Fi—No. 1 on Google when you search for “steal wi-fi”—consistently pulls in 250 hits a day. Jill Hunter Pellettieri’s 2005 defense of Rachael Ray (No. 5 for “Rachael Ray”) attracts 500-odd page views daily. And “One Giant Lift for Mankind,” Josh Levin’s 2004 piece about the 1,000-pound bench press, gets about 500 hits a week—which may have something to do with the fact that it’s at the top of Google’s list for “bench press 1000.” Of course, the trouble with letting readers romp through your archives is that you never know what they’ll root up. Another piece that consistently pulls 30 hits a week? A 2001 Vice column on a fabricated sport called “monkeyfishing” that we published (and then apologized for).

Slate editors also long ago devised ways to steer readers toward the old content we want them to read: “Recycled” (the department in which we flag archived pieces that become timely again) and “Related in Slate” (the little squibs at the bottom of each article with links to relevant archived material). When we recycled Negar Akhavi’s 2001 piece about Hezbollah last Monday, it got 737 hits. The day before, it had received only one. And that same day, when we linked to a 1996 piece about flight attendant English in a “Related in Slate,” the piece got 1,130 hits. It had received only two the day before.

How much of Slate’s traffic is generated by these less-popular pieces? Working with the traffic data for Tuesday, July 18 (which is about as much as my computer can crunch; click here for a note on my admittedly primitive methodology), I figured out that 8,903 articles attracted at least one page view that day. The 13 most popular articles accounted for 50 percent of Slate’s traffic, and the top 100 articles accounted for 86 percent. But our least-popular stories got plenty of notice. Seventeen percent of our traffic was generated by pieces that got fewer than 1,000 hits, 8.7 percent was generated by pieces that got fewer than 100, and just under 3 percent was earned by pieces that had fewer than 10. Graph that out, and you get a nice-looking tail.

So, it looks like Anderson can carve another notch in his laptop. What’s harder to pin down is whether Slate’s tail is pulling its weight. We know that somewhere around 9 percent of our traffic is generated by pieces that get fewer than 100 hits a day. But is that a lot or a little?

I e-mailed Anderson to ask how we could tell. “Depending on how you define ‘old’ (one day, one week, one month, etc), you could have as little as 20% or as much as 50% of traffic going to your archives,” he said. I decided to set the cut-off point at one week—the amount of time Slate keeps a link to each story on the home page. I took another pass at the data and found that just 22 percent of our traffic was generated by pieces that are more than a week old. That’s within Anderson’s cutoff, but barely. Perhaps Slate’s tail should be working harder.

How else could we lure readers into looking through our bric-a-brac? A few weeks ago, Jack Shafer noted that sometimes recommends archived Slate pieces—but his attempt to manipulate the site’s recommendation mechanism backfired. Wikipedia has a “random article generator”—the digital analogue to the experience of flipping through the encyclopedia. Perhaps Slate could offer a similar tool for browsing our back catalog. Or maybe we should replace the new module on our home page that tracks the most popular pieces on Slate with something else entirely, a tool that wouldn’t just promote articles that have already racked up plenty of hits: a Least-Read list.

Thanks to Daniel and Ben Engber for their help crunching Slate’s traffic data.