Twitter  Quartz found an unlikely inspiration for its relaunched homepage: The email newsletter.  
Nieman Journalism Lab
Pushing to the future of journalism — A project of the Nieman Foundation at Harvard

The newsonomics of paywalls all over the world

As more newspapers get on the paid-content bandwagon, there are a few promising models popping up. Here’s what to learn from them.

By the end of this year, figure that about 20 percent of the U.S.’s 1,400-plus dailies will be charging for digital access. Gannett’s February announcement that it’s going paywall at all its 80 newspapers galvanized attention; when the third largest U.S. newspaper site, the Los Angeles Times, went paid this week, more nodding was seen in publishers’ suites.

More than a dozen dailies in Europe are charging, led by Finland’s Sanoma (see “Sanoma’s Big Bundled Success”), Axel Springer, and News Corp.’s Times of London. It looks like more than a dozen in Germany alone may be charging by year’s end. In Asia, the powerful Singapore Press Holdings is first out of the gate, with other dailies there planning to follow.

Suddenly it’s paywalls all around the world. We’ve moved — in a couple of years — from the question of whether to when. The big question that should be asked now: How?

Charging for digital access is a nuanced question. For smart publishers, it’s part of a much larger strategic shift, touching every part of their operations: circulation, content, and advertising.

Let’s look at the newsonomics of an increasing paywalled world. The well-publicized New York Times digital scheme has gotten most of the attention, but it’s a global news source — more akin to The Wall Street Journal, the BBC, The Guardian, and CNN than to regional and local dailies.

While the Times is a fledgling pay model success, we can’t say, broadly, that paywall models are widely successful. Most aren’t failures, but few can point to the significant revenue difference that The New York Times, WSJ, or Financial Times plans have made to their transitioning businesses. Why? And what are the emerging successful formulas?

First, it should be said that the sky has neither opened up into a dazzling blue future nor fallen. A couple of years ago, predictions about the impact of paywalls mostly fell on the doomsday side of the equation. About the same time that going pay was proclaimed as another sign of the imminent death of Old Media, some were poking fun at the new iPad as a big smartphone that no one would want to hold up to her ear. Time to chill on the whole doomsday storytelling — we’re all in for lots more twists and turns.

So if charging for digital access — a too long phrase, but one that’s most accurate than paywall — is neither a panacea nor a tombstone on the way to the inevitable, what is it? It’s a building block, and it’s a way to re-envision the business.

It’s about a major shift in strategy, says Star Tribune publisher Mike Klingensmith, whose paper went pay on Nov. 1.

“We’re changing the nature of the customer relationship,” he told me. “Instead of the website undermining pricing of your content, it supports the pricing of your content” — seizing on the profound difference the all-access revolution is beginning to make. Relationships don’t change overnight, and that’s one important lesson to draw here: If newspaper companies can do more than offer lip service about relationship and “membership,” they have the ability to recreate an updated version of the trusted, community-oriented relationship that the better dailies long held. If they can reinvent the relationship, they have a shot at transforming themselves (“The newsonomics of crossover”) as they move into the mostly digital era.

Let’s look at some of the metrics learned from the early pay period, in talking with a number of the business executives who have been at the forefront of this grand experiment.

The big bogeyman of digital ad loss

The first big question that’s been laid to rest is the journalistic corollary of the Hippocratic Oath: Do no (advertising) harm. Remember the big fear about “going pay”: Would a paywall decrease digital visitors so much as to harm the only part of newspaper publishers’ businesses that’s growing, digital advertising? Metered models, like The New York Times’ (“At Almost 400,000 Digital Subscribers, Inside the New York Times Pay Strategy, Year 2″) and the Los Angeles Times’, are now the trade’s standard, having been advocated strongly by Press+ when it got rolling in 2009. Allowing 10-20 free articles a month has meant that traffic loss has been minimal; given the near-infinite amount of digital ad inventory, such traffic loss has had practically no effect on digital ad sales.

“All of the almost 300 publishers now using Press+ have kept their online ad revenues because we use data to make sure there is plenty of ad inventory to meet advertiser demand,” says co-founder Gordon Crovitz of Press+, which was acquired by RR Donnelley last year.

Even if some papers experience a small negative impact, new digital revenue quickly outpaces it. “In our first month of paid service, online subscription revenue was 3x the network advertising we lost because of the drop in pageviews, and our online subscription revenue has grown every month since,” says Andy Waters, general manager of the Columbia Daily Tribune in Missouri, which went pay on Dec. 1, 2010.

Pageview loss has ranged as high as 40 percent (at the Columbia Daily Tribune) and has typically run about 10-15 percent. Interestingly, from Minneapolis to Columbia to Hamburg, traffic often begins to grow markedly after the initial shock of a paywall. It may take months or a couple of years, but traffic is essentially reset and can then be rebuilt. Clearly, the most important readers — core readers who really use the news product through the week — have stayed the course.

The flipside of a tougher paywall is a higher signup rate, and more revenue, from those valuing the content.

Remove one major fear.

Selling more papers

One reason some papers went pay: Try to reduce the number of subscribers fleeing print. So far, there’s been a minimal impact on retaining subscribers, or “reducing churn,” as it is called in the business. The Memphis Commercial Appeal’s publisher Joe Pepe points to a 1 percent increase in Sunday home delivery, similar to what The New York Times has found. In Minneapolis, the Star Tribune has gotten 20 percent of its 15,000 “digital-only” subscribers to pony up an additional 29 cents (!) a week to get the Sunday Strib.

The Sunday sale is a major part of the how we see rolling out. At the Strib, it’s an inside-out, outside-in offer. If you only take the Sunday paper (subscribers who get two or more days of the paper delivered get free digital access), you’ll get a low, introductory rate to add digital access; if you’re a digital signup, you’ll be pitched on the 29-cent deal.

The L.A. Times is putting its own spin on the Sunday deal: pay 99 cents a week for the first four weeks (and $1.99 thereafter) to get free digital access and the Sunday paper. Want just free digital access only — that’ll cost $3.99 a week. You don’t have to be a coupon professional to figure out the better deal. The LAT approach mimics the NYT approach, which charges readers about $60 a year more if they refuse to take the Sunday paper. Maybe we should call it the Godfather offer.

How much will Sunday (“The newsonomics of Sunday paper/tablet subscriptions”) grow, given such pricing — which I expect more metros will adopt, given that they still have relatively weighty, ad-revenue-rich Sunday papers? The first job is to stop the Sunday bleeding, and if combined digital/Sunday products do it, consider it a tourniquet that publishers hope to get a couple of years out of, even as daily print circulation continues to decline. The Sunday angle — the Sunday paper angle — is a big one.

New money

While The New York Times is on a double-digit circulation (print + digital) revenue trajectory, other papers are having a hard time reaching that number. Columbia points to a 5-6 percent lift, enough to cover several newsroom positions for the small daily. Minneapolis points to a 3.75 percent lift, based on its new $1.5 million revenue stream, earned at $100 a year (or $2/week) from 15,000 digital subscribers. Others say the circulation revenue is flat to a little up.

One little secret of the trade: the opt-out. Build in higher pricing for combined print and digital access, and allow readers to take print only — if they affirmatively opt out. Eighty percent or so won’t opt out, and so we’ve seen high retention rates among newer subscribers.

The wild card here is how much the all-access offer — part of the changing customer relationship the Star Tribune’s Mike Klingensmith suggests — allows papers to price up their overall print/all-access subscriptions. He says the paper priced up its overall subscriptions 9 percent last spring, with little negative impact, the first time it had priced up in recent memory. Another increase is in order for this fall.

That’s the big key here, I think: If you tell customers “we’ll get you our content however, wherever you want it” — and deliver on that proposition with products that match the tablet and smartphone age — the creation of added value makes sense to readers. So it’s important to look beyond digital-only revenue itself, and look at the total reader-revenue-producing potential of smart pay plans.

As Gannett points to a goal of adding $100 million in new revenue, which would be a 10 percent circulation rev boost overall, look for as much of that to come from upward pricing in general as new digital-only subs themselves.

That said, it’s useful to pay attention to a new emerging metric: what percentage of a newspaper’s site unique visitors are signing up for digital access-only subs. The New York Times broke the 1 percent barrier last year, 390,000 subs compared to 33 million U.S. unique visitors. The Commercial Appeal is at .8 percent; The Star Tribune is at .25 percent with its four-month initiative. The Columbia Tribune is at .2 percent. It’s just one metric, but one that tells us about comparative traction. Though, it seems like a tiny number, it’s not. Fly-by traffic, supplied by Google and now Facebook, supplies so much traffic that about 3 percent of most newspaper sites’ unique visitors equal their paid print circulations. The digital-only conversion metric provides an apples-to-apples comparison, even as overall print/digital circulation impact remains key — and is measured in that old standby, dollars, euros, and pounds.

The goal here: Head to 50 percent of overall revenues being paid by readers.

These numbers are only a snapshot and come from some of the better practitioners of the digital pay craft. Many more are underachieving. The point is that there is an emerging playbook of how to get pay working right.

For now, let’s boil it down the how to 5 P’s:

  • People: As in customers. Few newspapers — probably a dozen or fewer in the U.S. — know their combined print and digital audiences as a single audience. It takes a lot of technology moving to get a single, whole view of a customer, matching the subscriber database with the digital registration database to get a holistic view. Without that view, it’s tough to operate a modern, somewhat digital/somewhat print business — and maximize the value of new pay propositions. The New York Times, the Star Tribune, and the Commercial Appeal are among those who do, and papers as small as The Day are getting there.
  • Product: This is a simple question of content. How much strong local coverage are readers missing after a half decade of staff cuts? The better a news organization covers its community, the more it can dare to charge and still get customer traction. Some papers may simply have already cut too much.
  • Presentation: Consumers — us — understand the all-access pitch. News (and magazine) publishers have to make it real. That means real ready-for-the-tablet (and smartphone) products, app-based and HTML5. Replica-plus products will satisfy paying readers less and less over time — and won’t compete with Flipboard-esque experiences.
  • Pricing: Enough said. Newspaper (and magazine) pricing has been fairly dumb over the years, a follow-the-leader, seat-of-the-pants exercise. Playing with the value equation, print and digital, requires both testing and matching of new value to new price.
  • Promotion: More than just marketing, the new promotion makes better psychological sense of the all-access proposition to older and newer (and younger) customers.

So 5 P’s — or maybe more.

“You have to do eight things right,” says Gregor Waller, a former exec at Axel Springer and now CEO of Digital Age Consulting, who is in the midst of advising a number of major media globally on pay models. “It’s like a golf swing. If you miss out on one, you can’t hit the ball correctly.”

What to read next
Ken Doctor    Aug. 25, 2014
America’s largest newspaper company says it’s building for the future. But it’s hurting its own value proposition in the process.
  • robelroy

    Good post, Ken. I might only point out that I rather prefer the digital replicas to the more fandangled versions. I think that Le Monde newspaper does it the best, with the paper laid out ‘as is’ and then you can tap on an article to blow up the text.

  • Anonymous

    I got a different offer from the LAT then the one Ken describes here. Mine says that I can pay 99 cents per week and get full digital access, or full digital access plus Sunday. Home delivery seven days a week costs $3.99 per week. Maybe the offers vary depending on where you live?  

  • Anonymous

    Never mind. I see now that was just for the initial four weeks.

  • Michael Conniff

    This is the best piece on digital paywalls I’ve ever read. Seems very nascent at this moment, but thank you for the digging.

  • Bob Mentzinger Jr.

    Traditional metered paywalls act as deterrent to interaction. Why not a reverse paywall: Pay a flat monthly all-access membership fee that declines to a base based on how many stories you read/how many clicks/how much time you interact with the product. Gets you revenues plus eyeballs, rather than punishing your most loyal customers.

  • Greg Golebiewski

    Many publishers claim that their subs figures are growing, but hardly anyone wants to elaborate on this growth or give any data.

    During the Paywall Strategies 2012 conference last month, only one publisher, Haymarket Media’s Autosport, reported that its paid subscribers represent about 1% of all its monthly visitors. Others, including The Economist and The Times refused to comment. However, based on the NYT 390,000 stat as the number of paying users at the end of 2011 and the overall pre-paywall traffic, it seems the paper has managed to convert only about 1% of its uniques. Are these results good, one or two users out of 100 visiting the sites?

    Given that the NYT reportedly spent $25M to erect its metered paywall and additional $18M to promote it, the cost of acquiring one subscriber was enormous, somewhere between $110-$155/sub.

  • Susan

    One thing that the papers haven’t figured out yet (or at least, haven’t figured how to deter), is that one can read unlimited numbers of articles simply by using a service such as Chrome’s “incognito” windows.  Using that service “masks” your IP, and you can usually read up to 15-20 articles before the site “recognizes” the temporary IP and asks you to subscribe.  At that point, simply close that cloaked browser, and open a new one to continue reading.  The downside is that, unless a reader does actually pay, they are unable to interact in the commenting section; however, if the primary intent is simply to read the article, then this is the way to do so.

  • Greg Golebiewski

     You should be ashamed of yourself Susan, so should Google, if indeed its Chrome allows for such a gimmick to steal content.

    How come that people are so proud of finding a way to cheat and even dare to instruct others how to do it?

  • Susan

    Actually, Internet Explorer has the same feature, it’s not exclusive to Chrome.

  • Amy C.

    Interesting post, but as a sociology grad student, I find it a little difficult to follow your train of thought.  Your title indicates “all over the world” but your stats and anecdotes are very US-centric…is this indeed representative of global news organization trends?  And did you take into account the growing presence of web-based news platforms, like Boston’s own GlobalPost?  I’d be curious to know your thoughts on how web-based news factors into the equation, as they don’t charge for access partly by virtue of lower overhead costs.

  • Yebostaki

    Still not sure that a pay model works except if you give me unique content. Digital copies also need to be quick to read wheels print can be loner versions.. Pay walls in the long run will not work if combined with advertising. One or the other not both.

  • Elliott F

    Interesting article. Have you done any research into how smaller newspapers are addressing these issues?

  • cas127

     “Are these results good, one or two users out of 100 visiting the sites?”

    And I think it is worse than that – what about churn?  How many of those alleged 390k new NYT subs are going to vanish (over and over again) once their “4 weeks for 99 cents” deal is over? 

    And of course the NYT’s pageviews haven’t been dramatically hurt – they didn’t put up a paywall – they put up a picket fence.  Working around the NYT paywall is so laughably easy that it may not even require actual effort – open a new tab, new window, set auto cookie erase, etc. 

    And the NYT sales force will just start selling on “reach” (uniques per month) rather than “frequency” (page views per month) anyway – but we’ll see how that latest media selling scam goes over…

    …Which brings up the LAT paywall (which appears a bit tougher than the NYT farce/fence/whatever).  I’ll bet that the LAT very shortly will *only* be reporting uniques and not pageviews.

    But the uniques will be magically “stable” – why?  Because they will count all those who hit the wall and get auto-directed to the front page.  Yeah…that’s valuable traffic…

    These “newspapers” committed to the “truth” are awfully opaque on the *specific* (as opposed to *selected*) results of their paywalls.

  • John Greenwood


    There is a clear need for a viable means of paying very small amounts for web access which avoids the disadvantages of paywalls and advertising.

    Paywalls are unpopular, the user has the bother of signing-on and remembering passwords. They are expensive to run and not practical for really small amounts, which means that the user is being charged a lot more than they feel for content is worth.

    There is a culture that all information should be free but in reality there are always costs. On the internet this is very small but not zero.

    If we take the example of Wikipedia where the content is provided free and the user is not charged. However there are hosting costs for the hardware, administration etc. According to the October 2011 report the total expenses for  September 2011 were $5.4M and the number of page requests were 15.8 billion, giving a cost of roughly 3000 pages/$.

    For Wikipedia this is raised by charitable giving. Currently the organisers find they able to do this easily. For most users it is only free because it is a gift.

    There must be countless other organisations that want to provide free content but have to find a way of paying for hosting. The alternatives are charity or advertising.

    I suggest that the option of paying about 0.1 ¢  would be widely welcomed and used, provided it involved the user with no more effort than a single click of a button.

    I would argue that this would make information more free as it can be made available without depending on charity or being acceptable to advertisers.

    The argument in support of advertising is that “There Is No Alternative”, followed by, “you can always just ignore it”. As well as being irritating to many users, (personally I hate it!) advertising takes up space on the screen, sometimes displacing the content off screen so that the user is has to do more actions to view what they actually want. On portable devices the advertisements can consume a significant extra amount of power, running down the battery.

    The alternative: W-stamps

    Web access is never free. The majority of users will be paying an ISP or a phone company for access. It makes sense to use them as the financial intermediary rather than invoke a new third party.

    The ISP will issue W-stamps to the user. This is exchanged for an item of content from the content host. The content host redeems the cost from the ISP. The ISP would adds the cost to users account. For most clients this will be a trivial addition to their regular fees and the ISP might offer several thousand pages of download as just one more service it is supplying.

    The administrative costs would be minimal. ISP adds the charge to the suppliers account together with all the charges arising from downloads by all its other clients from this supplier and makes periodical payments when the amount owed becomes significant.
    The mechanism of making the contract might be:

    * A new HTML markup token is agreed upon which contains the the identity of the page supplier and the charge.
    * When the user downloads a page containing this token, the download is paused.
    * The user’s browser displays a button displaying the amount charged. The user clicks on this button and thereby agrees to the charge and sends an electronic W-Stamp to the supplier whereupon the download is continued.

    * W-Stamps are issued to the user by their ISP in the form of a small data files containing the identity of the ISP and some encryption/key data.
    * Both the page supplier and the user  send modified copies of the W-Stamp back to the ISP with the transaction data encrypted/stamped in such a way that:

    ** the ISP knows that the request came from a W-Stamp that it originally issued

    ** the supplier knows that the user who requested the page is a client of the ISP.

    ** both agree on the page charge

    ** a malicious forth party cannot read or modify the details of the contract nor create a false one.

    Note that the only action the user has to make is one extra click. Even this could be automated by the user’s browser by giving automated agreements, perhaps up to some limit.

    Getting agreement, designing and getting widespread adoption of this mechanism will be a major undertaking and it would be a considerable time before it was ubiquitous. In the interim  users that have the properly equipped browser and participating ISP would be able pay via the above mechanism and get the requested page, while users not so equipped would get still get the page but preceded by an advert or a request for a donation.

    The normal W-Stamp is, like a postage stamp, essentially a convenient way of paying for transmission and would not give any rights to the content being transmitted. However the same payment mechanism could used to pay for content as well, an Extended W-Stamp. These would cost more than normal Web Stamps, perhaps up to a few cents. For example the reader of an online newspaper or books would encounter an Extended W-Stamp at the head of each article or section. The user would simply click to get the next bit. A charge of 1 ¢  for a section of material that takes about a minute to read would provide an hour of reading for $1, of the same order as the printed version. The difference for the reader would be that they do not have to buy a whole newspaper to read just a few articles. In the case of a novel, if they get bored after the first chapter they do not have to pay for what they no longer want to read.

  • Kyle Porter

    Imagine an all access platform where handfuls of publications are members and I get a single subscription where I can access all the sites but only get charged per each article view. You could easily build the framework & supply buyers with browser extensions to control the flow. Then it could store all the articles you read along with notes/highlights & indexing. Single service news campanion…pay per view..??