The NYT toughens up its paywall

By Felix Salmon
December 7, 2010
my advice when it comes to how to build a paywall.

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

Martin Niesenholtz, the head of digital at the NYT, clearly hasn’t been taking my advice when it comes to how to build a paywall. Instead, he’s pre-emptively cracking down on a tiny and financially meaningless minority of hypothetical readers who might want to find ways around his wall:

“We will take great pains to make sure that the first-click-free policy isn’t abused in any way,” said Martin Nisenholtz, the Times’ digital chief. “Google has been quite cooperative in terms of setting a limit for the number of free articles that can go in for any one day, so that you can’t just sit and engineer your way into a free use of the website.”

What this says to me is that the NYT is spending too much time designing its paywall, and is disappearing down rabbit-holes best left unexplored unless and until it becomes clear that they need examining.

The NYT‘s paywall is designed to be porous: readers coming in from some other site (Google, Twitter, Facebook, Reuters) will always be able to read the article they’re looking for, even if they’ve used up their monthly quota. As a result, it’s more of a navigation fee than a charge for content.

That’s fine—except Nisenholtz now seems to be backpedaling from that concept, and saying that he’s going to “take great pains” to crack down on people who read a lot of without paying the company.

That’s silly, for three reasons. Firstly, great pains tend to come at non-negligible expense, and there’s no point in spending significant amounts of money unless you think you’ll recoup those costs in extra revenues. In this case, Nisenholtz seems to think that (a) there will be a large number of people trying to find a way around the NYT paywall — and that (b) a significant proportion of those people will end up giving in and subscribing (as opposed to simply going elsewhere), if the paywall is made hard to get around. I very much doubt that he has any concrete evidence that either proposition is true, let alone that both of them are; common sense, then, would dictate that he wait until he gets such evidence before working on bolstering the wall.

Secondly, there are so many ways to get around paywalls—simply deleting your cookies generally does the trick—that there’s no good reason to believe the NYT‘s “great pains” are going to actually work very well in practice.

Finally, the less porous the wall, the more annoying it is—for subscribers and non-subscribers both. That’s simply the way that paywalls work. Strengthening your paywall sends the message that you don’t trust your subscribers, or your subscribers’ non-subscriber friends: you’re treating them as potential content thieves.

Why would Niesenholtz do this? Why won’t he just satisfy himself with raising revenue from loyal readers, rather than trying to prevent people from reading lots of stories? It should be flattering that some people want to read NYT content so badly that they will take the long way round the paywall. Instead, Nisenholtz seems to find it downright threatening. And one possible reason is hinted at by Rick Edmonds:

A Kindle subscription to the Times cost $19.99 a month, and Scott Heeken-Canedy, president of The New York Times newspaper, said that might be indicative of where pricing for full Web access will end up.

If by “will end up” he means “will end up eventually, after we’ve quietly raised the subscription price half a dozen times,” then Niesenholtz’s tactics don’t make sense. But if by “end up” he means “will end up being when the paywall goes live next year”, then they do.

$20 per month is a large amount of money for people to pay for a product they’re used to getting for free, and indeed it’s so large that most of the NYT‘s regular readers will simply refuse to pay it. In that situation, the subset of people who will pay but only if the paywall is a tough one might start becoming relevant.

Niesenholtz says that 15% of current visitors view 20 pages or more per month. But people won’t pay $20 to read 20 pages per month: that kind of money only begins to be worth paying once you start reading a few pages per day, or say 100 pages per month. Let’s say that 5% of current visitors fall into that bucket, and that of that 5%, only one in ten will actually pay $20 a month for website access. (I’m not counting print or iPad subscribers who get free access to the website with their other subscription.)

At that point, Niesenholtz is directly monetizing just 0.5% of his visitor base—a number which is small enough that grabbing a few people who otherwise like to find a way round paywalls could actually make a significant difference.

I think it would be silly to start the paywall experiment at a high $20-a-month price point. Somewhere between $5 and $10 would make more sense. But if the NYT really is considering making a $240-a-year product, then maybe that explains their newfound emphasis on cracking down on those who would try to get around it.


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

I won’t do a paywall at any price, not because I’m cheap or that I don’t appreciate content (content in one form or another is how I make my living), but because it’s too annoying and tedious to keep track of individual subscriptions to one-off websites. If I could have a single fund that gives my access to multiple paid sites of interest, I would almost certainly do so, but publishers are interested in being islands, not communities.

Posted by Curmudgeon | Report as abusive

Personally, I’m still awaiting my refund from the unused portion of TimesSelect.

Posted by Gorillameek | Report as abusive

I plan on purchasing an online subscription to, at pretty much any of the price points I’ve seen. I probably view 150-200 articles per month. I am 23 years old without much disposable income, but NYT is my primary source of news, and I value it quite highly. Above all else, I feel some kind of responsibility to ensure that a highly professional news organization like the NYT continues to exist, and for that reason they will definitely get a bit of leeway on prices.

Posted by efnj | Report as abusive

If the NYT is going to experience anything other than the 90% drop in access the London Times did it will need to be vastly different. There’s really nothing there I can’t get elsewhere. I have a copy of Keynes General Theory, so nothing Krugman says surprises; does anyone seriously think anyone is going to pay money to read Maureen Dowd? Lots of hubris for a cage-lining commodity content.

Now Reuters is doing interesting things, and I’m not saying that boost egos there. We have full contact blogging using Felix, financial Netflix with Chrystia, but most importantly an attempt to understand a new thing that is changing things at a cellular level by experimenting with new things. NYT looks like it is still trying to figure out how to make money without changing. Not sure what the future of journalism looks like, but I’m pretty sure that it doesn’t look like the NYT does today.

Posted by ARJTurgot2 | Report as abusive

I agree with your $5-$10 price point. I check out at least a dozen news sites or more every day including NYT. While I would pay $5 a month for a few sites, there’s no way I would pay $20/month for a dozen sites if paywalls become common.

I wonder if they tested the pricing model with the huge number of visitors they get every day or whether they just did a dartboard model to see what sticks.

Posted by SFGary | Report as abusive

Felix, Reuters should put you behind a paywall and put you on a 100% commission compensation plan.

Posted by eortiz | Report as abusive

A paywall is OK, but it needs to be properly priced. That is, modest. We are paying for content, not delivery, and anyone who is remotely familiar with the economic truths of media understands that circulation pays for circulation.

The actually delivery of Web content is minuscule when compared to home or newsstand delivery.Operating revenue is generated by ADVERTISING.

What the NYT needs is proof of demographics of readers so that they show those demographics and “page views” to advertisers. What is the difference if I read the ad in the paper or online? I am still exposed to the advertising. As an online reader, I should be rewarded with a lower price per “view” as I am saving the Times money as they do not have to deliver the paper to my doorstep (not to mention the waste of natural resources–and sad to say the demise of the “paperboy.”

Reasonable, even low, prices for online access should prevail in exchange, readers must submit to some demographic scrutiny.

Posted by eksommer | Report as abusive

$240 is toothpick money if you’re a partner at Goldman Sachs, but it’s a daunting sum if you’re nearing retirement on a nest egg more modest than those owned by New York’s uppercrusters. If the NYT thinks $20 a month is a modest amount for online access, it’s a bit out of touch with the real world.

But I don’t think that’s the problem. The NYT is a premium product, and its owners and managers think it should sport a premium price. That’s partly a strategy to protect the paper’s value, partly a strategy to protect the managers and owners from the humiliation of offering Wal Mart prices.

It’s also an old way of thinking. I’d price the paywall for unlimited access at a dollar a month, and work to get 30 million subscribers.

Posted by jrconner | Report as abusive

But in fact, Liu Jiangen this is not lost, his mind is persistent and firm, he wanted to go to Hengda, want to play, play in AFC Champions League, more bluntly, he want to get more money

hey, Hey make sure you assist me to affect the elgg, interpersonal torch theme to 1. main You should i want to recognize that site within mod to be able to alter to create my configuration with the active features Ough should create your individual layouts’. Ex., directly into folder views

Id should consult you here. Which isn’t the one thing I normally do! I like studying an article that should make folks think. Additionally, thanks for permitting me to comment!