Patching holes in paywalls, promoting journalism, calculating churn

By Grzegorz Piechota


Oxford, United Kingdom


Good morning! This is Readers First, a monthly newsletter for INMA members on reader revenue innovation. I’m Researcher-in-Residence at INMA. E-mail me at or DM via Slack (sign up here). 

Join me for a meet-up this morning to dig deeper into the topics discussed here: Tuesday, July 23, at 10:00 a.m. New York time. Register now. 

Was this e-mail forwarded to you? Sign up to this newsletter.


1. FREE-RIDING. How many holes are there in your news site’s paywall? 

Three of four news paywalls, or 75%, can be bypassed by cleaning cookies. Two thirds, or 68%, can be overridden by browsing in a private mode, finds a 2019 study by U.K. researchers from Imperial College London and Brave Software.

Methodology: For a test of bypassing techniques, researchers chose 25 paywalled sites, such as Bloomberg, The Irish Times, Medium, Le Parisien, The Spectator, and Wired. They browsed the sites until they were stopped by a paywall, and then they tested known bypassing techniques.

Search Google for “How to bypass a paywall” and you’ll get 100,000+ pages advising how to get around the payment on the leading news Web sites.

Developers of browser add-ons, such as Bypass Paywalls for Firefox, encourage users to submit proposals of their favourite sites to be hacked.

Reliability of paywalls: Overall, UK researchers were able to bypass all of the metered paywalled sites tested and none of the hard paywalls.

The likely reason is hard and freemium/premium paywalls restrict access on the server side and metered paywalls — the ones sampling content to users based on their past usage or propensity scoring — restrict access on the client side.

Bypassing techniques: It’s hard to say though how many news consumers out there do actually delete cookies or browse in a private mode to circumvent paywalls and avoid paying. The UK study did not survey users; it evaluated the sites only.

Earlier studies though indicate:

Counter measures: Many publishers and vendors have started actively patching the loopholes in their paywalls, for example, by using users’ hardware and software “fingerprints” instead of cookies and by detecting visits in a private mode. These efforts are disturbed by tech companies that zealously shift their approach to privacy. For example, in the latest version of Chrome browser, Google reportedly turns off the ability to detect a private mode.

2. ANALYSIS: To patch loopholes definitely, ask readers to register

Frustrated with the leaking paywalls, publishers will likely turn towards more restrictive sampling of content for free. The easiest way to patch all the loopholes is to allow access to some or most content and features to registered users only.

A registration wall may sound risky to publishers with significant reach and ad revenue, but it might also be a final call to shift from a browser-based digital marketing of the past to the people-based marketing of today.

The benefits for publishers of registrations and logging users in are plenty:

  • A unified customer identity.
  • View on single user behaviour across devices, browsers, and apps.
  • Ability to match behavioural data with other data sets, such as payment history.
  • Better segmentation, targeting, and personalisation of customer experiences — be it content recommendations or advertisements.

Marketing-wise, metered access was never about giving content for free indefinitely but about sampling of what’s paid.

The 2019 Imperial College/Brave Software study shows the free lunch is over across the world: Paywalls were found on 33% of sites in the global top-1,000 news index and 8% of the top global index of all sites. Out of 491 unique paywalled news sites studied, 67% used some sort of a meter. Expect the share of harder paywalls such as freemium/premium combinations (17% as of 2019) and hard paywalls (16%) to gain. 

How do you think publishers should respond to the free-riders? E-mail me:

3. TRENDS. News brands that promote journalism as a service instead of access to content

Story: Christine Vendel, an enterprise reporter from Harrisburg, Pennsylvania, starts work from home, giving her time to remind the kids to brush their teeth and walk the dog.

“But I spend a lot of my time out in the city investigating, interviewing, and researching. I want people to better know what’s happening in their community and we dig deeper to understand why,” she says in a June 2019 video ad of Advance Local’s

The slogan reads: “Christine Vendel asks the questions.”

What’s new here: This is one of the recent ad campaigns by online subscription-focused news organisations that shift from promoting their experience to credence attributes. 

Subscriptions to media content traditionally have been promoted as an experience good. Economists observed in the 1970s that utility of such goods is difficult to observe in advance without trying. Hence free samples, price discounts, and reviews were natural tactics for the media, similarly to restaurants, beauty salons, or travel companies.

Problem: In the age of online content abundance and fake news, readers may actually struggle to observe the quality of your content even after reading.

The value proposition to consumers by news organisations needs to be redefined, from providing access to content to the service of journalism based on witnessing, verifying, investigating, sense-making. These are traditional core roles of journalism, as defined by George Brock in his 2013 primer on news in digital age, Out of Print.

Solution: News organisations such as The New York Times, Advance Local’s PennLive of Harrisburg, Pennsylvania, or Poland’s Gazeta Wyborcza do not promote access to content any longer. Instead, they promote the process of journalism as an expert service.

Journalism can be defined as a credence good, similar in its essence to medical procedures, religious services, or organic food. Such goods are best marketed through reputation of the brand, expertise of employees, quality assurances and guarantees.

Further reading:

Inspiring examples of recent “credence” campaigns:

4. CASE STUDY. How Agora Radio monetises its podcasts 

Growth: Poland’s TOK FM, a news and talk radio channel, has more than 17,000 digital subscribers who pay US$2.60 to $US5 monthly for access to its podcasts. The number of subscribers has grown 60% since the last year.

Boom: In 38 markets surveyed by Reuters Institute for this year’s Digital News Report, every third adult, or 36%, accessed a podcast in the last month. The rise of podcasts got many news editors excited: Listeners are younger than readers; and although text still is a preferred format for news for most, users spend most their online time with audio, per eMarketer.

Engagement: “Surely, there will be fewer listeners on your Web site than readers,” says Jaroslaw Slizewski, chief digital officer of Agora Radio based in Warsaw and an INMA member, “but trust me — podcasts considerably increase the average time spent with your brand. A good podcast has more loyal listeners and receives more positive feedback than the best articles.”

Legacy: Thanks to its radio broadcasting operation, TOK FM is an audio factory. Every day, its 40+ journalists produce about 19 hours of new content for the broadcast and about two hours of new exclusive digital-only content: bespoke podcasts and extended versions of live shows. Add archives and subscribers get access to over 65,000 pieces of on-demand audio.

Publishing audio online actually requires a lot of text generation — each and every sound piece needs to be described and tagged. About 40% of content is also automatically transcribed.

Marketing: While TOK FM is present on Apple Podcasts and Spotify, it treats the platforms as marketing channels — distributing just clips and leading listeners to the full podcasts available only via the broadcaster’s Web site and the app.

A big benefit of TOK FM’s online subscription is an experience without audio ads. The cheapest package secures access to podcasts via the Web, while the more expensive ones offer access via an app and an RSS feed.

Lessons: Podcasts may engage users more than text, but getting users to listen to anything is also harder than making them read, according to Slizewski: “People value their attention.”

Podcasts are not attractive to all users and not all topics find their audience, so one perhaps should find a niche first and focus on a differentiated subject.

Consumption patterns may be different than with news texts, too: Some podcasts reach 50% of their total audience in the first two to three weeks; others wait months.

Want to learn more? Slizewski will be my guest at the next INMA online meet-up on Tuesday, July 23, at 10:00 a.m. New York time. Register now. 

5. INMA CLINIC. How to calculate churn? 

In theory, it’s simple: Churn is a ratio of a number of customers lost to a number at risk. In practice, there is no standard for calculating churn across news media. Churn rates are key to understand and manage health of relationship with customers.

Measuring churn vs. retention: Based on my experience, publishers in the United States tend to focus on retention (renewal) rates and publishers in Europe seem to use churn (defection) rates.

In theory, these are inverse metrics — 20% churn rate is an equivalent of 80% retention rate. But in practice there is a psychological difference.

Yasmin Namini, an ex-chief consumer officer of The New York Times, and some others believe the focus on churn rather than renewal makes you more serious about preventing any churn rather than complacent about your renewal rates.

Counting customers vs. revenue: Depending on the goal of the analysis, publishers can track the number of customers who end their relationship with the company or the revenue lost from these relationships. 

The former is useful to manage churn in product management or customer service; the latter is used in finance as an input to customer lifetime value calculations, projections of revenue, or cashflow.

By the way, revenue churn may be negative while your customer churn is positive — it happens when you manage to upsell or cross-sell the retained subscribers despite losing some less profitable ones.

Defining subscribers: Publishers differ in ways how they define “subscribers,” and that choice has an influence on what numbers do they use in calculations.

Some look at the total base of all people entitled to access the site. Others look at fully paid subscribers only, so they exclude people on discounted trials or sponsored subscriptions.

Another caveat is who is counted as a “new” subscriber excluded from the calculation. In theory, “new” should exclude subscribers who “renewed” their time-limited contract. Based on my experience, CRMs of some publishers don’t actually allow that.

Defining timeframe: Some publishers measure churn in “calendar time” such as weeks, months or years, and others choose “customer time” — cohorts of users surviving three, six, or 12 months since signing up.

Calculations at the segment level assume all individuals in a segment have the same probability to end the relationship. Other ways to analyse churn would be to look at churn by marketing campaigns, promotion, tier, type of a customer, etc.

This post has been inspired by an INMA member Dennis van Dongen of TMG Media/ Mediahuis, a company with activities in Belgium, Netherlands, and Ireland.

Further readings – reports:

Recent articles on tactics from the INMA members:

About this newsletter 

Today’s newsletter is written by Grzegorz (Greg) Piechota, Researcher-in-Residence at INMA, based in Oxford, England. Every month, I share results of my original research, notes from visits to digital subscription leaders, reflections on talks at conferences, and my favourite readings. Previous editions are archived online: 

This newsletter is a public face of a year-long reader revenue and media subscriptions initiative by INMA, outlined here. E-mail me at with thoughts, suggestions, and questions. Sign up to our Slack channel.


About Grzegorz Piechota

By continuing to browse or by clicking “ACCEPT,” you agree to the storing of cookies on your device to enhance your site experience. To learn more about how we use cookies, please see our privacy policy.