May 3, 2019 7:00 AM

We Launched a Paywall. It Worked! Mostly.

Here’s what we learned, including some surprises, in year one.

A little over a year ago, we introduced a paywall at WIRED. The idea, as I wrote back then, was largely about us. To start, we wanted to give ourselves stronger structural incentives to do great reporting. When your business depends on subscriptions, your economic success depends on publishing stuff your readers love—not just stuff they click. It’s good to align one’s economic and editorial imperatives! And by so doing, we knew we’d be guaranteeing writers, editors, and designers that no one would be asked to create clickbait crap of the kind all digital reporters dread. (Though, admittedly, we do publish some of that still.)

But the idea was also broader. At WIRED we genuinely believe that journalism as a whole needs to diversify its revenue streams. The advertising business has supported this business for decades—but digital advertising is unruly, unpredictable, and slowly being swallowed by the social media platforms. Paywalls aren’t for every publication, and it would be nice to live in a world in which every reader could access every idea for free. But, in general, paid content seems like the best bet to help this essential and embattled industry. So, with that spirit in mind, here are some thoughts about what we learned in year one that might apply to other publications.

First off: It worked! Of course you’d expect me to say that, but it really did. I promise. We increased the number of new digital subscribers in the first year by nearly 300 percent over the year before. We don’t know if they’ll resubscribe (please do); we don’t know if they’ll ultimately pay higher prices (please do); we don’t know if it’ll be as easy to get the next batch of people to join (please do). But the early signs are good, particularly for a year in which the bottom fell out from some traffic referrers that used to drive subscribers (hello, Facebook) and the greatest growth was on a platform (hello, Apple News) where getting direct subscribers in 2018 was as easy as hitting a bank shot 3-pointer, and getting subscribers in 2019 will now essentially require a half-court heave.

The second lesson: The stories that led people to subscribe were a little surprising. When we started this, we invested in three new kinds of pieces: longform reporting, Ideas essays, and issue guides. All three types overindex in generating subscriptions. But they weren’t the only things that drove subs. Here are the 11 pieces that drove the most subscriptions this past year.

1. Inside Facebook's Hellish Two Years—and Mark Zuckerberg's Struggle to Fix it All

2. The Genius Neuroscientist Who Might Hold the Key to True AI

3. Dr. Elon & Mr. Musk: Life Inside Tesla's Production Hell

4. Editors' Letter: The Next 25 Years of WIRED Start Today

5. Cyber Monday 2018: The Absolute Best Tech Deals Online

6. How to Set Up and Use a YubiKey for Online Security

7. Meet Jim Allison, the Carousing Texan Who Just Won a Nobel Prize for His Cancer Breakthrough

8. Pan Am Flight 103: Robert Mueller's 30-Year Search for Justice

9. A Complete Guide to All 17 (Known) Trump and Russia Investigations

10. Why It's Hard to Escape Amazon's Long Reach

11. Best After-Christmas Sales (2018): Bose, Beats, Blu-ray, and More

About half the stories are the kinds of pieces I would expect to lead to subscriptions: meaty features that got lots of traffic. But some, including #2, didn’t get that much readership. Others either weren’t features or weren’t quite so meaty. And it’s interesting that two of the top 11 were essentially buying guides—though very well-done ones. So what’s the lesson here? I think it’s that people will subscribe after reading all kinds of stories if they’re done well.

They’ll particularly pay, we also learned, if you send them newsletters. The propensity to subscribe by people who enter WIRED.com on a mobile device is rather low—unless they come in via a newsletter. (To give one data point, a visitor who reaches us via search is 1/19th as likely to subscribe as one who comes in from a newsletter; a reader coming in from Facebook is 1/12th; and a reader coming in from Twitter is 1/6th.) That’s one reason why we’re launching all kinds of new newsletters, tied to specific sections of the site.

Our experiences with gifts was odd. At the beginning of the year, we offered subscribers a new YubiKey. That was a great value! And it may have created a secondary market for YubiKeys on eBay. (It also, hilariously, might explain why story #6 on the list above did so well.) Later we tested giving people a free device to cover your laptop camera, which you really should do. Oddly, this depressed the rate at which people subscribed on the site and through email. They were less likely to pay the same price for a subscription if we gave them a camera cover. Why? I have no idea. Perhaps it made them feel creeped out. We also, at one point, tested a partnership offering a free short-term subscription to a partner brand that shall remain nameless. The result? Our response rate tanked. Perhaps because people worried that their credit cards would just get scooped up by someone else and auto-renewed into eternity.

As anyone who runs an online business knows, the order form is also extremely important, and we spent loads of time trying to optimize ours. We shortened text, refined the lines of boxes, and played with different images. Adding Amazon Pay as an option seemed to help a lot. But there’s still work to do. If you’ve got ideas for how we can make our order form better, let me know.

We also ran a bunch of interesting experiments. When we asked people to “place order” instead of “start my subscription,” 9 percent more did so. When we included coupons with the offer, fewer people subscribed than when we told them they could just “Save 50%.” For some reason, it seems, people much prefer “deliver to” and “customize your offer” over “choose a destination” or “choose an offer.” We also learned a lesson that every retailer ever knows: If you offer a sale, people buy. The question for us is whether those people will renew.

Those are just a few of the decisions we have worked through. And, of course, there’s more. WIRED is partnering with Apple for the launch of Apple News Plus, a subscription bundle that lets you subscribe to a bunch of publications at once. We’re excited by this—as long as it doesn’t cannibalize direct subscribers. You get much more WIRED when you actually subscribe to WIRED (archives, everything we publish online, an app, a beautiful magazine that the mail carrier will take to your door). If you’ve gotten this far in this particular wonky post, you’re the kind of person who probably knows that already. But we don’t know about everyone else.

What happens next? Well, in 2019 we’re going to double-down on all those things that we love to do, that we know people love to read, and that drive subscriptions. There will be more investigations, more guides, more gear. It seems like the president may make news this year, and I suspect we’ll write some more about Facebook too.

As I noted a year ago, there’s always going to be a tension between Stewart Brand’s famous dictum that “information wants to be free” and the lesser-known clause that followed: “But information also wants to be expensive.” The most important lesson, though, is the one we’ve learned over and over. Subscribers just really want information to be good.