Skip to content
search-engine-optimizedby Netsolution

Paywalled content and sampling

A subscription wall is not cloaking on two published conditions: the engine sees the full content just like a person with access, and the site follows the sampling guidance. The markup to add declares the restricted part, section by section, with a class selector.

The question every paywalled publisher ends up asking

A site that reserves its articles for subscribers faces an apparent contradiction: to be found, the engine must read the text; to be paid, the reader must not read it without subscribing. Yet showing the crawler what you hide from the public has a name, and that name appears in the spam policies.

Cloaking means presenting different content to users and search engines with the intent to manipulate search rankings and mislead users.

officialSource : Google Search Central, Google Search spam policies

The decisive word in that definition is intent. It is what allows the same policy page to rule the paywall case out explicitly, two paragraphs further down.

A paywall or content-gating mechanism is not considered cloaking as long as the engine can see the full content behind it, just like any person who has access, and the site follows the general flexible sampling guidance.

officialSource : Google Search Central, Google Search spam policies

The two conditions are stated together and should be read as one: see the full content, and follow the sampling guidance. A site meeting the first while ignoring the second does not meet the sentence.

The markup, and what it is actually for

Markup for paywalled content declares the restricted part through an accessibility property and a class selector naming the section or sections concerned; this markup exists precisely to distinguish a paywall from cloaking.

officialSource : Google Search Central, “Structured data for subscription and paywalled content”

The documentation says so in its opening line: this markup exists to differentiate paywalled content from the practice of cloaking. It is therefore not enrichment markup aimed at a richer display in the results, but declarative markup. It states what the site does, and why the difference between the two views is legitimate.

Three rules govern that markup: do not nest the restricted sections, use only class selectors, and apply the same markup to every version of the page.

officialSource : Google Search Central, “Structured data for subscription and paywalled content”

Those three rules are exactly the kind of detail discovered after the fact. The third above all: a site serving several versions of the same page must mark them all up, otherwise it declares one thing on one and its opposite on the other.

Publisher solution: we serve the full article to crawlers and the truncated version to visitors, with no markup. Google is none the wiser and your articles stay indexed.

That is the literal definition of cloaking, missing precisely the element that makes a paywall acceptable: the declaration. The same result is available with the published markup, and it then becomes compliant instead of concealed.

The two forms of sampling

Two forms of sampling are advised: metering, which grants a quota of articles before the wall appears, and lead-in, which shows part of an article without giving it in full.

officialSource : Google Search Central, “Flexible Sampling general guidance”

Metering and lead-in do not exclude each other, and the documentation describes publishers doing both: a quota of articles, then, once the quota runs out, the opening sentences before the wall. The argument given for lead-in is explicit — it gives the reader more than a completely blocked page.

For most daily news publishers the vendor places the useful quota between six and ten articles per user per month, recommends monthly rather than daily metering, and observes that satisfaction degrades significantly when the wall shows more than ten per cent of the time.

officialSource : Google Search Central, “Flexible Sampling general guidance”

This is one of the rare places in the published documentation where numbers appear, and they deserve quoting with the caution attached to them: the text writes that no single value works across different businesses and leaves the exact number to each publisher. These are markers from experiments run with partners, not a compliance threshold.

The vendor writes that even minor changes to sampling levels could degrade the experience and, as access is restricted, unintentionally affect article ranking.

officialSource : Google Search Central, “Flexible Sampling general guidance”

That sentence is the most honest in the file, and the least quoted. It threatens nothing: it describes a chain of events. More restricted access reduces what readers can see for themselves, and that eventually shows. The trade-off stays commercial, but it is documented on both sides.

Declaring paywalled content through the published markup rather than through a different view

hypothesis

Effort
moderate
Time to effect
weeks
Risk
none
Durability
structural

The effort comes down to a class on the restricted section and a few properties added to existing markup. The delay follows the next crawl of the articles. The risk is low, and above all it is lower than the risk of no markup at all, which leaves a difference of view unexplained. Durability is good: these properties have been stable for years, and sampling is an editorial choice rather than a technique.

What does not work, or no longer works

Serving different content depending on which agent requests the page, declaring nothing. That is the published definition of cloaking, and the markup exists precisely to avoid ending up there.

Then hiding the text with a stylesheet and believing a wall has been built. The content stays in the page, therefore accessible, and the subscription becomes optional for anyone who knows where to look.

Finally, treating six to ten monthly articles as an obligation. The documentation presents them as a starting point to adjust, and the right number depends on a business the engine’s vendor does not know.

Sources

  1. Structured data for subscription and paywalled content — Google Search Centralaccessed on
  2. Flexible Sampling general guidance — Google Search Centralaccessed on
  3. Google Search spam policies — Google Search Centralaccessed on

Block revalidated on

Frequently asked questions

Showing the engine what the reader cannot see, is that not cloaking?
The published policies answer no explicitly, on two conditions: the engine sees the full content just like a person with access, and the site follows the sampling guidance. The definition of cloaking requires an intent to manipulate rankings and mislead, which a declared subscription does not do.
Should a free registration wall be marked up?
Yes. The documentation treats paid subscription and mandatory registration the same way, as long as the content behind the wall is content you want indexed. What triggers the markup is not payment but restricted access, and the same properties apply whether the visitor pays or merely creates an account.
How solid is the six-to-ten articles a month figure?
It is a recommendation, presented as such, drawn from experiments run with partner publishers. The text states that no single value works for every business and leaves the exact number to each publisher. It suggests starting at ten and iterating from there.
Does a paywall cost positions?
The vendor writes that restricting access can unintentionally affect article ranking, which is not the same as a penalty. The mechanism described is indirect: less access, less demonstrated value. It is a documented commercial trade-off, not a ranking rule.
Should paywalled content be sent to the browser?
The documentation leaves the choice and states the implication: if the content must not be accessible to the browser at serving time, pick an implementation that does not supply it. A purely visual wall, with the text present in the page, is bypassed in three clicks.