How the NYT should construct its paywall

By Felix Salmon
January 18, 2010
Gabriel Sherman says that the NYT is going to be putting up an FT-style paywall:

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

Gabriel Sherman says that the NYT is going to be putting up an FT-style paywall:

After a year of sometimes fraught debate inside the paper, the choice for some time has been between a Wall Street Journal-type pay wall and the metered system adopted by the Financial Times, in which readers can sample a certain number of free articles before being asked to subscribe. The Times seems to have settled on the metered system.

Let’s hope that if and when this happens, the NYT implements a good metered system, rather than the bad system adopted by the FT. The NYT arguably has the best news website in the world right now, and any kind of paywall implementation is going to result in a deterioration of the experience of using it. But if the NYT is smart about this, it can try its hardest to be be as unobtrusive and user-friendly as possible.

The first and most important principle that the NYT must bear in mind is that any smart metering system will work more like a taxicab than like the dreadful FT approach: the key thing is that a meter measures how much of the service you’ve consumed, and then you pay for that much — and no more. At the FT, by contrast, the meter slams down a hard paywall after you’ve reached n pageviews in a given month, and then charges you a very large sum for the n+1th pageview. That’s stupid, because no single pageview is worth that much to a reader.

The NYT system should instead simply measure how much you used the site last month, and then bill you; my guess is that Apple, when it releases its new tablet later this month, will also unveil a system which makes it very easy to link your account to your iTunes account so that your NYT bill will simply get added on to your iTunes bill along with your apps and TV shows and music and ringtones. The NYT itself won’t even need to collect your credit-card information. Once you reach a certain maximum billing level for the year, the NYT and Apple will just stop billing you.

And what of people who can’t or won’t pay? The NYT is an invaluable source of information for many people around the world who don’t have credit cards or iTunes accounts. I think that in the first instance most countries outside the US (and maybe also Canada) should be exempt from having to pay anything. There will always be ways of getting around any paywall, so the NYT shouldn’t worry too much about loopholes; its first priority should be retaining as much of the user experience for as many of its readers as possible, given that it has come to the conclusion that a paywall of some description is necessary.

In fact, I don’t think that the NYT should ever bar content entirely from people who can’t or won’t pay. If you refuse to be billed for usage, then the NYT can maybe massively increase the number of interstitials you get served, or otherwise allow the kind of intrusive advertising that normally it shuns. (Click to make this survey go away!) The Sulzbergers are rightly serious about preserving the NYT as a public trust — and it’s in their self-interest to do so. Nowadays, that public trust is best served through the website, and its accessibility to hundreds of millions of information-hungry people around the world. Cutting them off from the website just for the sake of dealing with a nasty cashflow problem is not smart business. As I said about the FT last year, high-profile newspapers should value long-term brand value, and the ability to stay relevant to future generations of potential readers, at least as much as short-term subscription revenues.

I trust too that NYT digital chief Martin Nisenholtz will do his best to encourage bloggers to keep on linking to, just as he came to an agreement with Dave Winer, before he freed up the NYT’s archives, to create a way of linking to stories whereby the links would continue to work even after the story should have disappeared behind the archive wall. One easy and obvious policy would simply be to say that the first story you see after following a link to from some other site is always free. If purpose-built websites started popping up which existed only to allow people to get around the paywall very easily, the NYT could always exempt those sites from the policy.

More generally, the NYT should build the easiest and most user-friendly metering system they can construct, without worrying too much about whether and how their readers are going to be able to game the system. One of the many problems with the FT’s system, in particular, is that it’s overly paranoid, to the point at which it regularly blocks paid subscribers from the site. But the fact is that there will always be people trying to game the system — and they will always be in the minority. If and when that minority becomes very large, the NYT can revisit its paywall design. But in the first instance, it shouldn’t worry too much about them. It’s the same idea as spam filters and comment moderation on blogs: they should be implemented only after comment spam becomes a problem, not before.

For the NYT has bigger things to worry about than readers gaming its paywall. (Remember that it doesn’t worry about people stealing copies of its physical newspaper — and those extra copies cost real money to print and distribute.) Up until now, the NYT has been in a very small group of news sources, along with Reuters, the BBC, and the Guardian, which people know that they can link to, safe in the knowledge that their link is going to work forever: anybody clicking on the link will get the story. A badly-designed paywall will oust the NYT from that select group — which would be great for Reuters, but very very bad indeed for the NYT. Let’s hope that if and when this paywall arrives, it’s designed with all the intelligence and conscientiousness that we currently see informing the paper’s site design.


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

“At the FT, by contrast, the meter slams down a hard paywall after you’ve reached n pageviews in a given month, and then charges you a very large sum for the n+1th pageview. That’s stupid, because no single pageview is worth that much to a reader.”

The FT system worked on me. I eventually succumbed, even though I get the print version as well. (website access is extra, which is a bummer)

However, the performance of is pathetic. It can take 20 seconds for the home page to load. Not a patch on NYT or WSJ.

Posted by BasabPradhan | Report as abusive

It’s hard to discriminate between different kinds of links. Would a “linked articles are free” policy work with links from Google or Wikipedia? What if you bookmark an article and then go back to it? What if you go to an article from a cached link?

I think that, to be technically robust, discrimination between links has to as simple as possible and on the server side. So, e.g., front-page, day-of-publication articles are free (which makes sense because they are the most attractive to advertisers), but everything else is micro-metered.

Posted by MattF | Report as abusive

I think any pay-per-article meter system is doomed. Who wants to keep track of a dozen different meters at different media sites?

I think the newspaper sites should be allowed an antitrust exemption to form consortiums. Say you combined the NYTimes, the WashPost, papers from Europe, Asia, etc., and then charged $150/year and divided the proceeds based on page views.

Posted by Bob_in_MA | Report as abusive

The FT paywall is in fact completely porous. The metering works via a cookie — when you reach your quota, just delete the cookie and get a fresh quota. It’s even easier than defeating the WSJ paywall, which is done by googling the article title.

Posted by Statman | Report as abusive

Ummmm….Which cookies do you delete on the site to bypass the firewall?

I just use rotating throw away email addresses, but deleting a specific cookie would be more convenient.

Posted by Matthew_Saroff | Report as abusive

What’s really amazing, is that both the method to circumvent the WSJ’s paywell and that of the FT are simple bugs. I dealt with that in 1999, it isn’t rocket science.

The main reason newspapers haven’t been able to make money on their Web sites is that they are run by idiots. I remember reading 3-4 years ago that the NYT’s Web site was bragging that they had reached break-even. But that was only if one assumed that all the content they received was free! If they had given me the contents of the NYT, I would have been profitable within a week.

Posted by Bob_in_MA | Report as abusive

Newspapers have been giving away content for free to readers for generations.

What the readers have paid for is a portion of the cost of putting ink on paper.

Advertisers pay for content.

The problem in newspapers is two fold:
* The bean counters have been systematically cutting the quality of the product to make quarterly numbers.
* Their most profitable section, classifieds, is having its lunch eaten by Craigs List, etc.

Posted by Matthew_Saroff | Report as abusive

Internet content is not free. I pay about $60/month for my internet content. The problem is that ALL the money is going to the distributor (service provider). Granted I’m paying for high speed pipes, but I certainly would care for those pipes if there was nothing to view, read or hear. Certainly, this ”meter” way of calculating internet use is fair, but they should collect the insanely rich providers.

Why can’t Murdoch and friends take their fight elsewhere?

Posted by maxturbo | Report as abusive

You suggest some very good models but how about just a flat amount, say $25 a year for people who use the site a lot and free to its paper subscriber? By a lot I mean 30-50 page views every day. Or as a commenter responding to the Jarvis/Buzzmachine diatribe suggested the Pandora model of a nominal sum of 99cents/month supported by ads and a full access @ $35 w/o ads.

I do agree that they should develop a model that is painless for its userbase. I am a big fan but if they start locking content like FT or WSJ and go beyond my dollar threshold I’ll go elsewhere even though there’s very few reliable national news sites in the U.S.

Posted by SFGary | Report as abusive