What Is a Paywall? How They Work, Types & Examples

You are reading an article and in the third paragraph it doesn’t make any more sense. A gray box appears on the page: “Subscribe to keep reading. That box is a paywall, and whether you’re on the side of the subscribe button or not makes you either a big fan of how someone gets paid for their work, or the number one bane of reading news on your screen.

Both of those things can be true at once. Paywalls have moved from a few newspapers trying out a brave new idea in the early 2010s to being the standard setting for news sites, newsletters, online courses, podcasts and even whole paid communities! If you do any kind of reading on the Internet on a regular basis, you’ve hit one of these. As a website owner, blogger or newsletter publisher, you have most likely considered whether or not you should build one.

Here’s a pragmatic examination of what a paywall really is, how it works, the various types of paywalls, and, most guides fail to mention this, what effect a paywall actually has on a site’s relationship with Google.

What a paywall really is

The rule is applied automatically: if someone has paid or signed in, display this content, otherwise, either display a preview, a locked message, or even a blank space where the article used to be. It’s not something in the flesh on a server. It’s logic. A PHP code determines who is requesting the page and determines what they can see of it.

That rule shows up in more places than most people realize. The obvious example is news publishers like The Guardian, The Economist and The Wall Street Journal, but the same basic process is in play with newsletters on Substack and Beehiiv, courses on Teachable, communities on Discord and Circle, and podcasts with a bonus feed for paying listeners. Wherever a person is trying to make their audience into direct income, there’s a good chance that there’s a paywall, or something close enough to it to be considered a paywall, doing the work.

There is a lot of slack usage around, though; it’s good to be precise. With a paywall, content is limited. The payment relationship is typically a subscription that lies behind it. People use them interchangeably, but you can have one and not the other; a single article can be sold without a content wall subscription, and a content wall subscription can include perks, such as early access or a private newsletter, or ad-free video — none of which have anything to do with the content wall itself.

What’s actually happening behind the screen

Here’s the part most explanations skip. Each time someone loads a page, a paywall must resolve three questions: who is free and who is gated, has the visitor at this instance earned access to the paywall, and if the answer is no, how much does the visitor get to see before the paywall falls.

The first question is typically determined ahead of time by an editor or a rule: breaking news remains open, more in-depth features get gated, etc. The second is the place where technical work takes place. Websites that follow a metered model usually place a cookie or log a browser fingerprint the first time that you appear on any website, and then increment a counter each time that you open a new article. After passing the monthly limit, the counter reverts to the wall. Other sites that require a log-in don’t even bother with the guesswork: no account, no access. This is a better way of doing it, but it creates friction at a cost to some of the readers.

Technically, there also is a difference in gating location. Others load the whole article into your browser and then use JavaScript or CSS to obscure content after a certain point, which may not be possible to circumvent by someone who wants to read the article. Others don’t even send the gated text to your browser; even the server will determine what to send you first. The second one is more difficult to overcome, but it also has a real impact on the way that search engines treat the page, and I’ll be returning to this later, but it is commonly the one that catches out publishers.

In any case, the meter usually resets on a rolling monthly basis, not on a daily one. It’s not a coincidence. Google’s own publisher guidance is to use monthly cycles instead of daily ones, as it is believed a monthly reset will allow readers to actually experience the contents of the publication before deciding to pay for it, but not so generous that no one will ever choose to pay.

The different types, and who actually uses them

Paywalls are not one design with various logos. The only thing they’re really different about is how much a publisher is willing to give away before requesting payment, which affects who the model is a good fit for.

Hard paywalls

A hard paywall is a wall that displays a headline and possibly a sentence or two, and then simply stops. There will be no grace period, no sampling. Wall Street Journal is the most often cited one, it now has over four million digital subscribers, which required a commitment that there wasn’t a free alternative good enough to replace the financial reporting. That is what is needed for a true hard paywall. Not only good content, but content that a reader couldn’t find in another place anywhere for free. It’s an assertive approach, and a dangerous one to take for anyone without a niche that’s been built up and is unlikely to be defeated by any other model.

Metered (soft) paywall

A metered paywall is a system where a user can read a certain number of articles, typically between 4 and 10, before the paywall kicks in. This model has been in operation since 2011 and has resulted in this as the most successful news subscription model in print publishing history, with 12.5 million digital-only subscribers in early 2026 alone. The obvious appeal is that readers are able to try the work before committing to it, thus keeping search and social traffic on the site alive and still converting the loyal readers who are the ones visiting most often.

Freemium

Freemium breaks content into two infinite categories rather than a rolling counter: Some content will always be free, and a separate, clearly demarcated category is for paying members. There is a version of this on Medium, and in a lot of the newsletter and blogosphere, a free weekly newsletter is accompanied by an archive of members-only posts, or a series of posts with more depth. It is ideal for publishers that have a broad top-of-funnel audience and would like a front door that is always open.

Registration walls

This one is typically categorized under freemium, but it’s important to have its own category since no money is changing hands. A registration wall requesting an e-mail address and a free account rather than a payment. This is the system for most “sign up to keep reading” messages, and it is so popular for a reason: it’s easy and doesn’t require the reader to pull out a card in order to get access, and it provides the publisher with a first-party contact which can be used for future communications. Many sites have a registration wall as their initial gateway and then have a paid wall on top for those that sign up and continue to return.

Dynamic paywalls

A dynamic paywall does not employ a set figure at all. It evaluates each visitor’s behavior how often he visits, how far down he reads, where the traffic came from, etc. and adjusts the wall as it happens based on how likely the visitor is to actually subscribe. Though the textbook example of a hard-paywall, there is also a dynamic mechanism working under the surface of The Wall Street Journal that determines the level of friction on a case-by-case basis for each article and each reader. A more advanced setup and typically only really makes sense once a publisher has sufficient traffic and data science support that makes the modelling worthwhile.

Micropayments and pay-per-article

The model that occurs least frequently and is often called “about to take off” without taking off completely, because the user is charged a small sum for one item, but not for signing up to a subscription. There have been a number of platforms that have sought to make this a habit over the years, but with varying success. That friction of a new purchase decision at every individual article purchase is more difficult to work around than some kind of subscription that just quietly goes away. Today, it’s more likely to be an available option within a larger membership model, rather than a standalone business model.

Why publishers bother with any of this

Back in the day, advertising was the only thing, and with a lot of the internet, it remains the only thing. But ad rates fluctuate, and so do ad blocker penetration rates, and a page view by a non-returning reader isn’t worth a lot in comparison to a monthly paying reader who reads one article or 30 articles. Some publishers take a middle-ground approach, favoring the more native style of advertising, such as ads that look more like an in-text article than anything else, as opposed to any gating.

There’s another reason that tends to be overlooked: having a relationship with the reader. A publisher receives a direct connection to an individual who subscribes, or even registers, without relying on the algorithm of a given platform or a tracking system within an ad network. This has proved to be more significant than initially thought. For years, Google has been indicating that it planned to deprecate third-party cookies in its browser, Chrome, but backed off in 2024, which saw a reduction in the replacement technology it had been developing. However, Safari and Firefox do default to blocking third-party tracking, and even if Chrome decides to do so, a significant portion of any audience was never trackable this way. Those who were able to have a direct, logged-in relationship with readers avoided that whole mess. That’s the best business argument for a paywall or registration wall today aside from the subscription revenue.

There’s a lesser impact in addition to that. The incentive isn’t to get clicks anymore, it’s to write something that the reader will come back for. This does not necessarily mean that content that is paywalled is better. Plenty of it isn’t. It does, however, change the focus of what publishers are optimizing for, and for some, that’s enough in itself.

The trade-offs nobody fully avoids

None of this comes without cost. A paywall must have a real purpose; readers will typically bail if they encounter a wall with no good justification for it, and with free alternatives available to them elsewhere. It’s a very real possibility for anyone who doesn’t have a specific niche. It’s also a growth problem. Many young publications begin with a fully open front door, but close it down later on the moment they are able to discover what it is their readers want.

On the reader’s side, the complaints are just as real. Information that should remain public is behind a paywall, especially if it is breaking news, public health, or civic issues. Not everyone has the money to subscribe to five or six outlets to be read regularly. Placing a wall awkwardly is more likely to lead to workaround sites and browser extensions designed to remove paywalls altogether, which is not a good thing for any reader or for the reader’s privacy and security; the latter often comes with its own package.

Truthfully, a paywall is neither good nor bad, as an entity. It is a wager that a publisher will lose fewer viewers gating a piece of content than it is worth, a bet that is won or lost by different publishers on a regular basis. Of note, a few popular sites have dropped paywalls altogether, after determining that the trade was no longer worthwhile a reminder that it’s not something that just gets set and forgotten.

Is it worth trying to get around one?

If you Google “bypass paywall”, you’ll find a myriad of hacks: incognito windows, URL tweaks, browser extensions specifically designed to do this. None of it is a secret, and saying that it exists is not saying that you should use it.

The better question to ask is whether it is worth the effort, as there is likely another solution that is obvious. There are many publishers that offer free trials, student discounts or regional rates, or a few giveaway links from an existing subscriber. Most local libraries have services, which many cardholders don’t know, that allow them to read major newspapers for free. But if the story is worth sharing and worth reading, and you don’t want to circumnavigate the wall each time to get the original content rather than someone else’s paywalled version, paying for it directly is probably the more enduring solution. It’s someone’s got to pay for the reporting on which you rely, and advertising is not paying for it, usually.

If you want to create one on your own

It’s a matter of two honest questions: What is there that you can give away for free and still for which people will want to pay, and what do you have that is worth saving up for? A blog or a news-type site with a broad reach tends to fare better with a metered and/or registration model, as that is what is necessary for keeping the site discoverable while gradually converting those who are most interested. If your newsletter, course, or community is more targeted and has a smaller audience but is more engaged, you can often get away with a harder split freemium, or even more restrictive wall because you know when they are coming that they are there for a reason.

On a related note, you should consider what you’re gating and for what purpose. A publisher developing exclusive content for paying members must have that info feel distinctive, it can’t be precisely the same information with a lock symbol placed on it, or the wall feels like a tax. Regardless of the type you choose, begin easily less than what you think is comfortable. It’s much easier to make a wall tighter when you have a good idea of how people actually act than after you’ve made the wall too tight and readers have already decided you’re too aggressive.

The SEO side almost nobody explains properly

The part that surprises even some established site owners, and that it’s important to understand whether you ever intend to run a paywall yourself or not, is that it demonstrates a pattern that’s mentioned in numerous publisher SEO tips but with little context.

Google will NOT by default block gated content from ranking. The rule for years was First Click Free: Publishers were required to allow visitors from Google results to read at least one full article before they reached the wall. That requirement was dropped as of 2017 and is now Google’s flexible sampling. As is now the case, it’s up to the publisher whether or not to expose any information to anonymous visitors or crawlers, and the publisher can specify the gated information using structured data. The two that Google recommends are metering, the same free article allowance mentioned above, and lead-in, which is a part of an article that is always exposed, but not the full article.

The actual structured data is relatively easy to use: a boolean attribute isAccessibleForFree, and a hasPart element that references a CSS selector indicating precisely where the gated content starts. It’s something like this on a page’s JSON-LD:

{
  "@context": "https://schema.org",
  "@type": "Article",
  "isAccessibleForFree": false,
  "hasPart": {
    "@type": "WebPageElement",
    "isAccessibleForFree": false,
    "cssSelector": ".paywalled-content"
  }
}

That is because there is a problem it is there to solve, and that is that otherwise there is no way for Google to distinguish between a legitimate paywall and cloaking serving different content to Googlebot than to real visitors. Cloaking is a serious offense against Google’s spam policies, and it can lead to a page being demoted or not included in the Google index at all. The structured data, in a sense, is the publisher’s equivalent of saying, We know that there’s a difference between what the crawler sees and what the visitor sees, and we’re putting it on purpose; it’s not a trick to get you to rank higher. If you get that right, your gated page is indexed just like any other page. Or create a paywall that discreetly blocks text from actual users, but not from crawling visitors, and this website is risking a manual action rather than a ranking.

It’s a representative of a larger trend: search engines are excellent at rewarding clarity, and punishing sites that aren’t transparent. If you have any of these types of sites you’re running, it’s a good idea to read more about writing SEO friendly content in general because, beyond paywalls, transparency is very important.

The tools people actually use to build one

If you’re not simply reading about paywalls, things have come a long way since the days of hiring a developer to install a paywall in six weeks. Substack and Beehiiv are designed to create a paywall without any extra configurations. Ghost is a favorite for anyone looking to be in charge of their site but not want to stack on another platform, as the subscription and payment system are integrated natively. Both Memberful and Patreon function best as an overlay to a current website or viewers, not a complete platform change. The opposite end of the spectrum is Piano.io, a platform that’s designed for larger publishers with more complex metering, dynamic pricing and subscription logic, and is not necessary for most smaller sites.

For WordPress users in particular, a membership plugin that interfaces with Stripe will give the majority of a metered or enlistment wall, without a full move. It’s really a matter of what can be done to have control instead of just what can be handed to a platform that’s already implemented the payment logic and access logic for you.

Where this leaves the readers and the publishers

Paywalls are not going anywhere and, in fact, are proliferating in new areas such as newsletters, courses, community sites, and beyond that were not even part of the journalism landscape 10 years ago. For readers, the trick is to know which of the few subscriptions are worth paying for, and as for the rest, use the legitimate alternatives, the trials, and the library access, while students use the student pricing. For anyone who has a site, the sensible thing is to be honest about what is worth gating, choose the model which fits the size and loyalty of the audience you have today, not the one you wish to have, and if search traffic is important to you at all get structured data right from the start, rather than learning the hard way what works and what doesn’t.