Paywalled content markup: telling Google what sits behind a paywall

JSON-LD example

Copy this into a <script type="application/ld+json"> tag and replace the values with your own.

{
  "@context": "https://schema.org",
  "@type": "NewsArticle",
  "headline": "Inside the county budget: where the new transit money goes",
  "datePublished": "2026-09-21T06:00-05:00",
  "author": {
    "@type": "Person",
    "name": "Luis Ortega"
  },
  "isAccessibleForFree": false,
  "hasPart": {
    "@type": "WebPageElement",
    "isAccessibleForFree": false,
    "cssSelector": ".members-only"
  }
}

Properties

Is this for your site?

The markup matters only when two things are true: you want the locked text crawled and indexed, and Googlebot can actually fetch it. That usually means news publishers, subscription newsletters and paid learning sites with a flexible sampling approach.

If your membership plugin removes locked content from the HTML for everyone who isn't logged in, Googlebot sees only the teaser, just like a visitor. Nothing is hidden from users that Google can see, so there is no cloaking concern and the markup adds little. The same goes for content behind a hard login you would rather keep out of search.

Giving Googlebot the full text means recognizing it. User-agent strings are easy to fake, so verify requests the way Google recommends, with a reverse DNS lookup or its published IP ranges; see Googlebot.

Class names, arrays and nesting

Wrap each locked section in an element with a class, and reference that class in cssSelector:

<p>The first two paragraphs are free to everyone.</p>
<div class="members-only">
  <p>Subscriber text starts here...</p>
</div>

Several locked blocks on one page become an array of WebPageElement items, each with its own selector. Google's rules are short: only .class selectors, no nesting of locked sections inside other locked sections, and the same markup on every version of the page, AMP and non-AMP included. JSON-LD and microdata are both accepted. The property works on CreativeWork and several subtypes: Article, NewsArticle, Blog, Comment, Course, HowTo, Message, Review and WebPage.

JavaScript paywalls and WordPress membership plugins

Client-side paywalls send the full article in the HTML and hide it with a script or overlay. Google added guidance in August 2025 noting that this design makes it hard for it to work out what is paywalled and what isn't. If you use one, the markup is the clearest signal you can give. The better fix is a paywall that doesn't send locked text to the browser at all.

On WordPress, add the property to the Article markup Hydrogen SEO already prints, using the hydrogen_seo_schema_markup filter and whatever check your membership plugin provides:

add_filter( 'hydrogen_seo_schema_markup', function ( $schema, $post_id, $schema_type ) {
    if ( 'Article' === $schema_type && my_site_post_is_members_only( $post_id ) ) {
        $schema['isAccessibleForFree'] = false;
        $schema['hasPart'] = [
            '@type'               => 'WebPageElement',
            'isAccessibleForFree' => false,
            'cssSelector'         => '.members-only',
        ];
    }
    return $schema;
}, 10, 3 );

Replace my_site_post_is_members_only() with your plugin's own function, and make sure the class in the markup matches the wrapper your plugin outputs.

Controlling snippets and AI previews

Indexing the locked text doesn't mean you want it quoted. Google lists AI Overviews and AI Mode among the features that follow Search's preview controls, so the same tools apply there. Wrap passages in data-nosnippet to keep them out of snippets, or cap snippet length with max-snippet. Test both with URL Inspection in Search Console, which shows how Googlebot renders the page.