How to build a paywall

By Felix Salmon
July 14, 2010
this is a very sensible way to do it:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

If you’re going to put a paywall on your website, this is a very sensible way to do it:

Press+ is aware that there are ways users can avoid paying for the content its affiliates are about to charge for, including by using this Firefox extension (as well as by using multiple browsers), if they are willing to spend the time and effort and endure the related inconvenience.

As we develop the system in the coming months we will implement our plans to address this problem to the degree it is material for any publisher. But we’ll do so from the prospective that we are talking about content that has been free for years; thus the fact that a small percentage of people may try to circumvent a modest charge for it and succeed in doing so for a short period of time must be seen in light of the fact that most won’t, which means that a new revenue stream from loyal readers will have been created.

David Brauer seems to be of the opinion that any new paywall should be “robust” and shouldn’t be able to be defeated by means of a plugin (or by using multiple browsers, or by deleting cookies, or various other methods, I suppose). But that’s exactly wrong. The purpose of a paywall isn’t to keep people out, it’s to generate revenue from loyal readers. And the expense of making the paywall harder to circumvent is almost certainly greater than the marginal extra revenue that such an action would generate: after all, the kind of people trying to get around the paywall will most likely simply go elsewhere, rather than pay.

Back in the old days of print newspapers, you could read them for free by going to your local library or your nearest hotel lobby. You want a free copy of the FT? Just pop in to the FT building at 1330 Sixth Avenue — there’s always a pile there. A paywall is like the cover price on a newspaper: it’s the amount that you’re asking and expecting your readers to pay, but it’s not a sum without which reading the paper is impossible.

LancasterOnline is asking its readers to pay it $1.99 a month if they read a lot of obituaries. That’s a reasonable request and if I was a regular reader of that site’s obituaries, I would pay the fee. To sneak around the paywall is to place oneself in an antagonistic relationship with the paper you’re reading: it might be legal, but it’s certainly impolite. It’s a bit like stealing fruit from a roadside farmstand operating on the honor system. And I can’t imagine that’s the kind of thing that LancasterOnline’s loyal obituary readers would do.

Meanwhile, high paywalls impose other costs. When I’m reading Twitter on my phone and follow a link to the WSJ or FT, I hate running into their paywall. And the FT, in particular, has a paywall which breaks in unexpected and annoying ways, barring you from reading stories even when you’re a logged in subscriber. Recently a very generous multimillionaire, a huge admirer of the FT and a loyal subscriber of many years, told me that he hates the FT paywall — partly because he runs into it sometimes and partly because it makes it harder for him to share FT stories he likes. It’s pretty obviously not good business for the FT to alienate its customers like that and I suspect that part of the reason that the NYT is taking so long building its paywall is to try to avoid those problems as much as possible.

Circumventability, then, is a sign of a sensible paywall, not a sign of a badly-designed one. People will get around any paywall if they really want — just pasting the headline into Google News will often work fine. Newspapers shouldn’t worry about the people who do that; instead, they should be flattered. And spend their efforts instead on improving their relationships with their paying customers.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
8 comments so far

For sites that work with headlines from Google News, there’s an even easier way of skipping past paywalls. The way those sites work is by checking if the HTTP referrer is Google. A browser extension like RefControl for Firefox lets you specify the value of the “Referer” (sic) field in HTTP request.

So, I have nytimes.com, ft.com, wsj.com all set to use http://news.google.com/ as the referrer. I never hit their registration / paywalls, and instead go through to the article.

Posted by BarryKelly | Report as abusive

The subscriber link-sharing problem is fixable. When a paid-up subscriber views an article they get a “share this article” link, which is a custom URL specific to them and that article. Then the site imposes a maximum number of page views on that custom URL.

Posted by impedant | Report as abusive

(I meant to say also: the rest of the problems with the FT site losing subscriber cookies/etc are just really bad implementation problems)

Posted by impedant | Report as abusive

Felix – thanks for the name-check and very gentle disagreement.

I agree; paying up on some level is a matter of honor that loyal readers should recognize. JO’s shoplifting analogy is apt. However, such practices are more common online than in candy stores, I’d wager, and most retail outlets take minimal loss-prevention approaches, at least.

I think publishers should expect some minimum level of protection (though of course, that’s up to them). I don’t honestly know how expensive it is to defeat the JavaScript workaround – it might be quite easy.

Like you, I’d prefer a cooperative model, and maybe you’re right that more cooperation, not enforcement, is the way to go. Then again, I was gnashing my teeth yesterday at not being able to read ESPN Insider’s coverage of the NBA free agent follies … and almost considered paying up. The price was too high, but the solid paywall made me at least think about it.

Posted by dbrauer | Report as abusive

Is it me, or do the current attempts at charging for content seem even more lame than those of a decade and more ago? Back before the turn of the century, it was at least excusable to be clumsy about it, because it was new. There is no excuse for what we are seeing today.

I will pay for content, and I believe many people will. But I *will not* set up separate accounts on the several dozen sites that I might wish to read, thus greatly multiplying my chances of having my credit card number stolen, receiving junk email, and contending with numerous subscription renewals. Give me a single account that is usable across many different content sites, and decrement that account based on what I read on those sites.

To those of you seeking to get people to pay for content, is this so difficult to understand? I recognize that you have other motives (you desperately want my name, demographics, and contact info for yourself, for example), but what you are doing today will get you nothing.

Posted by Curmudgeon | Report as abusive

I grew up in Detroit and read those papers online. I would pay a little for access, because I am not going to see the physical paper unless I make a special trip for it (and thus a rational recurring cost would be a sensible economic choice for me). I’ve never understood why papers don’t charge out of area readers for access. I’ve also never understood why they couldn’t get in front of classified advertising – especially the huge paper groups – so I’m not genuinely surprised. If you’ve ever used Craigslist, you know it could be improved on manyfold and yet papers have, if anything, even worse systems.

Then again, I just tried the Boston Globe “Reader” app for a few days and canceled it because it was just plain lousy. It not only had authentication issues but proved harder to use than the full fledged website even on a small screen device. And it didn’t give me access to local ads so the experience relied completely on my paying for the app, which seems illogically disconnected from the way that newspapers carry ads in regular business.

You’d think that after so many years an industry that relies on information would have a better handle on some of these points but they don’t and that is their fault, not the internet’s.

Posted by jomiku | Report as abusive

Why bother with a paywall at all? The Miami Herald came up with a rather unique alternative for monetizing its operation that might be a model for other news organizations to consider.

Posted by politicalcalcs | Report as abusive

The paywall is very stupid. I don’t have a credit card, and never will. I pay cash for everything, or write a check. I don’t ever want to be in 25% interest hell. So I cannot access those sites with a paywall. But this is okay, because anything worthwhile gets cut and pasted to one of the sites I read for free. Talk about cutting of one’s nose to spite his face!

Posted by TaxLawyer | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/