Frederic Filloux at The Monday Note argues that the metered paywall approach can have substantial benefits for papers that implement one, as the New York Times has. But those positives are more than outweighed by the negatives, including the opportunity that paywalls create for free competitors.


Media industry research shows that more and more newspapers are implementing paywalls, possibly inspired by the launch of a “metered” wall at the New York Times earlier this year. Frederic Filloux argues in a blog post at The Monday Note that the metered approach can have benefits for papers that implement it, by boosting revenue and appealing to advertisers. But those positives can be more than outweighed by the negatives of a paywall, particularly for smaller newspapers — the main one being that a wall creates an opportunity for free competitors, of which there are a growing number.

In his post, Filloux makes the case that a well-constructed metered wall such as the one the New York Times launched in March — which was based in part on the experiences at the Financial Times‘ similar metered approach launched in 2007 — can produce revenue from a paper’s most devoted readers, and can do so without having much of an impact on advertising revenue. Even though page views inevitably decline with the launch of a wall, Filloux says most news sites don’t sell all of their advertising inventory anyway, and the loss of those eyeballs is made up for by subscription revenue.

The amount of inventory sold to advertisers varies widely. In the U.S. market, the “sell-trough” [sic] ratio is about 60 percent, but it can go as low as 30 percent on some markets. This means the media can sustain some loss in page views due to the implementation of the metered system without losing ad revenue.

Filloux theorizes that a site getting 5 million unique visitors and about 100 million page views per month likely has advertising revenue of about $15 million a year. If it can convince even 2 percent of its regular readers to pay $10 a month — which is similar to the approach taken by the New York Times — it could generate $10 million in additional income without eating into its existing ad revenue. In some cases, Filloux says, advertisers are even willing to pay more to advertise to readers who have subscribed, because they are seen as more engaged and therefore more valuable.

This is clearly the kind of math that the New York Times is hoping will work in its favor. A month or so after the paywall launched, the newspaper said it had racked up about 100,000 new subscribers, and NYT executives have said it hopes to increase that figure to about 300,000 by the end of the year. Despite those rosy numbers, however, even an optimistic view of the paywall’s financial outcome produces only $35 million or so in revenue — a drop in the bucket for a media company whose overall revenues are more than $500 million. According to Reuters media blogger Felix Salmon:

I hear that the brass at the New York Times expect its paywall to be revenue neutral — the amount of money they expect to bring in from online subscriptions is pretty much equal to the amount of money they expect to lose from online advertising.

Despite the math, a recent survey of the newspaper industry done by the Missouri School of Journalism found almost half of the small papers surveyed said they were implementing a paywall of some kind. Why do this if the numbers are so inconclusive? In most cases, these walls are likely to be driven by the same rationale that Rupert Murdoch used in launching paywalls at two of his British newspapers: namely, to keep print readers from deserting the paper product in favor of reading online, something that would remove a paper’s main source of ad revenue.

But to me, the biggest flaw in a paywall isn’t that the math is questionable, or even that a wall is inherently a backward-facing strategy, aimed at stacking sandbags around a paper’s content to try to keep out the digital hordes. The biggest flaw from a business perspective, particularly for smaller newspapers, is that walling up your content is an invitation to free competitors — from AOL’s Patch.com and Huffington Post to Mainstreet Connect and Neighborhoodr and Topix.net — to come and take away your readers.

Newspapers like the Financial Times and the Wall Street Journal  can make paywalls work because their content is extremely focused and (arguably) more valuable than that produced by free competitors. The New York Times is hoping it falls into that category as well, although as a mass-market newspaper, that conclusion is more of a gamble. But if you are a small-town or even medium-sized metro paper, walling off your content could be a recipe for disaster, by giving your more nimble competitors exactly what they are looking for: readers eager for a free alternative.

Post and thumbnail photos courtesy of Flickr user Giuseppe Bognanni

You’re subscribed! If you like, you can update your settings

  1. So there’s my home computer, where I use Firefox and Chrome, and my computer at the office, where I also use the same two browsers (and if I need, there’s Safari). Then I have a xoom, with 3 browsers, and an ipad, with 2. My android phone has a couple of browsers. I just haven’t come close to using up my allotted page views on their paywall, and that’s without trying to distribute my reading of the NYT over the various devices. I realize the number of devices I have is not typical, but a desktop, tablet, and phone is not unusual, nor is people using some combination of Firefox, Chrome, IE, and Safari. I was planning on getting the minimum print subscription that would give me complete digital access, but I haven’t hit the wall yet, so I still haven’t given the TImes any money. I just don’t think the paywall is going to matter for many people.

  2. Carl Morris Monday, June 6, 2011

    Filloux theorizes that a site getting 5 million unique visitors and about 100 million page views per month likely has advertising revenue of about $15 million a year.

    Is this a pro rata thing? That is, does 1m uniques and 20m views per month, say, result in £3m a year? I guess not and the advertising economies are different right?

    1. It’s difficult to extrapolate from a single site to smaller or larger sites, yes — partly because advertising rates are based on all kinds of other factors in addition to just raw pageviews or unique visitors.

  3. Mukesh Aggarwal Monday, June 6, 2011

    I seriously doubt this statement..

    ‘In some cases, Filloux says, advertisers are even willing to pay more to advertise to readers who have subscribed, because they are seen as more engaged and therefore more valuable.’

    If you were an advertiser, would you pay same or more amount of money for 2% of regular users who are more ‘engaged’ because they pay for news ? in addition people who pay would have low tolerance for ads anyways. not sure where Filloux got his data.
    Given that conversion rates for online ads is less than single digits numbers it is a no brainer what advertisers would pick.

    1. It is actually true – and always has been. In the pre-Internet era, advertisers paid more if a magazine had a high proportion of subscribers, because the publisher could tell them more precise information about who the readership were.

      A good example is MacUser UK, which I used to edit. MU was something like 70% subscription based, which meant that we knew we were hitting a very tight demographic of affluent graphic designers, largely paying via their company. One of our competitors at the time, MacFormat, had a far large sales number – but mostly via news-stand, which meant they never knew who their readers really were (they could guess, but that was all it was – guesswork and the odd survey). Because of this, our per-page revenue was far, far higher.

      (As an aside, this is why the ownership of the subscriber database is so important, and why publishers are fighting Apple over it. A subscriber who you know nothing about isn’t a benefit, in the sense that you can’t tell advertisers any demographic info about your subscriber base. In our case, had Apple owned the subscriber database, we would never have been able to tell advertisers that, say, 80% of our readers were freelance graphic designers.)

      So, now, with paywalls: Basically, if you don’t have one, you have lots and lots of traffic about which you know next to nothing. That makes each reader much less valuable to advertisers. With a paywall, you gather info about your subscribers that makes them more valuable.

      Take, for example, The Times (the London one!). I subscribe to the online version (not print – and I think Mathew is wrong about Rupe’s paywall being all about saving print subbers). They know who I am, where I live, what my job was.They also know what competitions I’ve entered and probably which sections I browse regularly. All of this can be used to show me better targetted – and hence more expensive – ads. That makes my eyeballs MUCH more valuable than some unknown person arriving on a story via Google.

      1. Thanks for the comment, Ian — and the info on MacUser.

      2. Very interesting. I would agree with you generally, advertisers like the demographics. However, would they would really pick 2 targeted users over 100 untrageted users ?
        Can you give more details when you say ‘our per-page revenue was far, far higher’. The way it reads it is , you were making lot more advertising money than your free competitor in absolute numbers (even considering that they had lot more free customers than your subscription based content). Is that correct ?

  4. Good points about paywalls increasing the threat of competition, though I did find the paywalled GigaOM related content at the end of the post ironic.

    1. Thanks, Kevin — I’m not saying that publishers can never charge for any kind of content, only that they should be very careful about what kinds they are charging for and the unintended consequences of doing so. Thanks for the comment.

  5. Alex Schleber Tuesday, June 7, 2011

    Spot on.

    I keep saying that the share of the Attention Pie in one’s ecosystem is the only thing that matters anymore, anything else is working under the illusion that they will still be able to charge for entry-level content:

    “Only then will some in the #dinomedia come to see, that the race was not about who was still going to eek out some residual “crumbs” profits from the Old System, but who was going to wholesale import the masses into their Ecosystem…

    Instead of dumb ideas like these: http://alexschleber.amplify.com/2011/04/22/new-york-times-dinomedia-upheaval/ , that only prove the deep denial that many from the Old Guard still find themselves in, because… well… the good old days, they were so very nice…

    While they lasted.”

    Some companies like Amazon (Kindle, etc.) and now Apple with iTunes Match/iCloud are understanding this. It’s ONLY about the ecosystem anymore. Murdoch and the NYT (ironic, isn’t it?), not so much…

    1. Alex, eyeballs that don’t make you money are pointless for a publishing company. Publishing is a business. Attention without monetisation is a cost, not a benefit.

      It’s ironic that you cite Amazon and Apple, two companies that understand this perfectly. Amazon doesn’t allow you to take its content and resell it, or even copy and paste from it. Apple… well, Apple’s attempts at lock-in and lock-down are well chronicled, as is its aversion to advertising.

      1. Sorry, but you are wrong. Attention imported into one’s ecosystem is THE key at this point. No attention = no profit. Period.

        Why? Because he knows that you have to build the Ecosystem first, then there will be plenty of opportunities to monetize later (compare vendors at a fair or sports event).

        Fast Company appears to be agreeing with me BTW: “The Apple Land Grab: How The Free Cloud Will Help Grow A Nation” http://www.fastcompany.com/1757782/the-apple-land-grab-there-is-such-thing-as-a-free-icloud

  6. “Filloux theorizes that a site getting 5 million unique visitors and about 100 million page views per month likely has advertising revenue of about $15 million a year. If it can convince even 2 percent of its regular readers to pay $10 a month — which is similar to the approach taken by the New York Times — it could generate $10 million”

    Those numbers don’t work.
    2% of 5,000,000 is 100,000.
    100,000 paying $10 per month is $1,000,000 per month.
    $1,000,000 per month is $12,000,000 per year.

    Not $10,000,000

    But, it’s not 2% of uniques. It’s 2% of “regular readers.”
    So, the question becomes what does Filloux assume for the percentage of uniques who are “regular readers.”
    Well, working backwards to get to $10,000,000 per year revenue..

    $10,000,000 per year revenue is $833,333.33 per month.
    At $10 per month, that’s 83,333 paid subscribers.
    83,333 is 2% of 4,166,167.

    Further, 4,166,167 “regulars” divided by 5,000,000 uniques
    is a whopping 83.3%

    I’m no newspaper industry insider, but I assert that if 83% of uniques to a website were “regular readers” the newspaper industry would be in a very different financial position these days.

    Curious about the source and nature of what seem obvious error(s) took me to the original post by Filloux.

    His thought experiment (not theory) is clearly laid out here:
    “Let’s try back-of-the-envelope calculations. A site gets 5m UVs and 100m page views per month; its yearly ARPU (Average Revenue Per User) coming from advertising is $3. This results in a yearly revenue of $15m. Now suppose only 20% of its audience reads more than 15 stories a month and one out of ten such readers are willing to pay $10 a month. The additional revenue will be: 5m UVs x 20% hitting the paywall x 10% willing to pay $100/year (discount included) = $10m in additional income — without depleting its advertising revenue. Actually, experience shows advertisers are now paying roughly 30% more for readers reached behind a paywall. All this before the 20% cut taken by Press+.”

    Clearly, 20% of uniques are “regular readers” and 10% of regular readers pay $10 per month. That gets you to 100,000 people paying a (unexplained discounted. pay for year in advance, I suppose) $100 per year… equals $10,000,000 per year.

    Mystery solved.

    To Paywall or Not to Paywall is nothing more complex than the old unique value proposition game.


    1. Thanks for correcting my math, Dan — I confess that I got a little lost in the intricacies of Filloux’s algorithm.

  7. Gordon Crovitz Tuesday, June 7, 2011

    The great thing about debates such as about the paid subscription model online is that ultimately the data determine the right answer. Dozens of the sites, including local newspapers, are now using Press+ (disclosure: where I’m co-founder) to implement “freemium”-style meters asking only the most frequent, enagaged online readers to pay for unlimited access. The results are in, consistent with Frederic Filloux’s Monday Note summary: All these sites retained all their online ad revenues and online readers month to month as they begin to convert 5%-10% of their online visitors to paying subscribers. Several have put in place ad rate increases (typically a 30% CPM premium for pages viewed by paying subscribers). These subscription revenues are key to building more successful business models for all forms of journalism, including both for print publications and online-only sites.

  8. K.M. Hurley Tuesday, June 7, 2011

    It seems to me that small newspapers are having so much trouble staying afloat that they “have to try something.” Or so they say. Another aspect is that often smaller papers are actually owned by News Corp (owner of WSJ, also going paywall) or similar. So they’re just falling into line. I resist pay walls. To be honest it’s mostly because I do not need another ten accounts to manage, even if (or especially if?) it’s $10 here, $6 there, $100 over here. Not worth the administrative hassle.

  9. jeremy bayston Tuesday, June 7, 2011

    I think there are a couple of assumptions here. Firstly, newspaper sites are relatively cheap because they don’t have to pay for content. To produce a cost neutral site when you have acres of content to put on it is actually quite easy. Secondly, I am not sure that the advertising models have been made clear here. If the models are PPC, or even CPM, then it would be difficult I think to state clearly that advertisers are paying more for those behind the paywall. They may well be paying more per lead generated, but not necessarily more in total. the danger, I think, for smaller newspapers is that they will disappear if they go behind a paywall. They won’t have a big budget, or sibling media to encourage people to pay up. Essentially the big papers want personal info, so that they can profile and segment their readership for advertisers. Smaller papers can do this more effectively by encouraging engagement with their readers so that they harvest this information with out having to go behind a paywall. Fb, Twitter and web interaction can encourage readers to hand over valuable info in a much more relaxed way than filling out a credit card form. It just needs people to think through a sensible strategy.

    1. Yes, what you’re saying, in effect, is that there is no one-size-fits-all business model for online newspapers. The advertising models vary drastically from publisher to publisher, where the big boys mostly sell display on a sponsorship basis and use CPM and CPC models largely for remnant, the medium sized publishers lacking a good sales team use CPM and CPC, and the small publishers are stuck with low rent CPC ads.

      Trouble is, the paywall assumptions are predicated on the big publishers’ top visitors (many sessions per month -> lots of ad impressions) opting to convert to paid. Hence the metered approach, which is a risk-mitigation scheme that allows the big publisher to continue to capture 20 page views (aka ~ 40 ad impressions a month) from their top segment, with–presumably–no direct effect on lower value segments who don’t hit that 20 page views/mo threshold.

      However, because that top segment is responsible for such a huge share of ad revenue, quite a bit of risk remains that a large portion of this core audience fails to convert to paid for any number of reasons, resulting in cannibalization of the most significant ad revenue stream…

      Anyway I basically agree with the blogger in that only the most premium content brands (read: national/international brands like FT, HBO, BBC) are sufficiently differentiated from the consumer’s perspective to justify a monthly subscription.

      1. Have there been any studies on whether regular readers click or or notice ads more or less than one-off readers who come in from links? If the latter are less inured to a site’s ads, a metered paywall may result in a smaller than expected drop in ad income. Another consequence would be that sites should instead be charging ad-premiums for NON-subscribers (in cases where no accurate subscriber personal data is available for ad targeting and selling.)

        It can also be the case that ad-monetization is inversely proportional to content quality. If a site with poor content can lure views through search engine results, readers will find that the most interesting things on a page are the ads, and so will interact with them more compared to a site where the content holds the reader’s attention.

  10. The paywall for news will notwork for so many reasons, not the least of which is there are so many free options. I find the odd thing is that most publishers do not know what business they are in. Advertising is becoming a less effective intermediary between audience and brand. Publishers need to have their content consumed on advertiser sites. Disclosure: This is what we do with video content at Vidsense.- Jaffer

Comments have been disabled for this post