WebSite schema markup (schema.org/WebSite) is a lightweight but often overlooked type that formally identifies a website as an entity with a name and canonical URL. Its most practical application is enabling the sitelinks search box: a live search field Google can display beneath your domain in branded SERPs, letting users search your site without leaving Google. You place this markup on the homepage only — it is a site-level declaration, not a per-page type.

On its own, WebSite markup is one of the simpler schema types to implement. What makes it worth prioritizing is its role in the entity stack: combined with Organization and Person markup, it rounds out the identity layer that tells Google exactly what your site is, who runs it, and what users can find there. I implement it on every new site build as part of the baseline structured data setup.

What WebSite Schema Markup Does

The primary functional benefit of WebSite schema is unlocking the sitelinks search box feature. When Google sees a valid WebSite block with a properly formed SearchAction on a homepage, it may display an internal search bar directly in branded search results. Users who already know your brand can search your content without clicking through first — reducing the chance competitors intercept that click.

Beyond the search box, WebSite markup feeds the knowledge graph. When Google finds a consistent WebSite entity that references the same Organization or Person described elsewhere on the site, it strengthens the overall identity signal and reduces entity disambiguation uncertainty. For sites with large content libraries, e-commerce stores, or membership platforms, the search box feature can noticeably reduce click-through to competitors after a branded search.

One thing to be clear about: schema is not a direct ranking factor, and the sitelinks search box is not guaranteed. It is a Google-controlled feature — the markup makes you eligible, but Google decides whether to display it based on your site's brand search volume, the functionality of your internal search, and other signals. Sites with substantial branded query traffic are most likely to see it activate.

Required and Recommended Properties

The WebSite type has a small property set compared to Organization or Person. These are the relevant fields:

  • @type — WebSite
  • name — the human-readable site name as you want it displayed in search features
  • url — canonical homepage URL; include or omit the trailing slash consistently with your canonical form
  • potentialAction — a SearchAction object required to trigger the sitelinks search box
  • SearchAction > target — an EntryPoint whose urlTemplate points to your search results URL with the query parameter templated as {search_term_string}
  • SearchAction > query-input — set to required name=search_term_string
  • publisher — a reference to your Organization or Person entity; optional but reinforces entity relationships and is worth including
  • description — a short summary of what the site covers; optional but feeds AI retrieval context

JSON-LD Example for WebSite Schema Markup

This block goes in the <head> of the homepage only. The SearchAction points to a WordPress native search URL — adjust the urlTemplate to match your CMS's search parameter if you use a different platform.

<script type='application/ld+json'>
{
  "@context": "https://schema.org",
  "@type": "WebSite",
  "name": "Terry Samuels",
  "url": "https://www.terrysamuels.com/",
  "description": "SEO consulting, schema markup, and technical SEO resources from a 14-year practitioner.",
  "publisher": {
    "@type": "Person",
    "name": "Terry Samuels",
    "url": "https://www.terrysamuels.com"
  },
  "potentialAction": {
    "@type": "SearchAction",
    "target": {
      "@type": "EntryPoint",
      "urlTemplate": "https://www.terrysamuels.com/?s={search_term_string}"
    },
    "query-input": "required name=search_term_string"
  }
}
</script>

Note on the target format: the schema.org spec accepts both a bare URL string and an EntryPoint object. Google's current documentation examples use the EntryPoint object form, which is what I use here. If you inherited markup that uses the bare string format, it will likely still work, but updating it to the object form future-proofs the implementation.

How to Implement WebSite Schema in WordPress

WordPress generates partial WebSite markup through Yoast SEO and Rank Math, but the defaults often omit the publisher reference and occasionally use a generic target string instead of the EntryPoint object. The simplest reliable approach is a custom JSON-LD snippet via the wp_head action, gated by is_front_page() so it only fires on the homepage.

If you prefer to use a plugin, check the output with the Google Rich Results Test after enabling it — Yoast's WebSite output format changes between major versions and the current form may differ from what their documentation shows. I cover both the plugin and manual approaches in the guide on how to add schema markup.

For platforms other than WordPress: Squarespace, Wix, and Webflow do not expose a built-in JSON-LD injection point with the same granularity. On those platforms, adding the WebSite block typically requires injecting custom HTML into the <head> via each platform's "custom code" settings. See the relevant platform-specific guides for details.

How to Test and Validate WebSite Schema

After implementing, validate with two tools:

  • Google Rich Results Test (search.google.com/test/rich-results) — shows the parsed entity and will flag if the SearchAction is malformed or the urlTemplate uses an unsupported format
  • Schema Markup Validator (validator.schema.org) — checks the full block against the schema.org vocabulary and will catch missing or incorrectly typed properties

If the sitelinks search box does not appear in branded SERPs after implementation, the most common reasons are: low brand search volume (Google sets a threshold before activating the feature), a urlTemplate that does not return functional search results, or a site that was recently launched and has not built enough branded query history. See the full guide on testing schema markup for a diagnostic checklist.

WebSite Schema and the Sitelinks Search Box in Practice

The sitelinks search box matters most for sites where users have a habit of going directly to the homepage and then searching within the site. E-commerce stores, large content publishers, and SaaS platforms with rich help documentation see the most measurable impact. For a small consultant site with 50 pages, the benefit is more about entity completeness than the search box itself.

I recommend implementing it regardless of site size because the cost is minimal — it's a handful of lines of JSON-LD on one page — and the entity signal benefit compounds with the Organization schema markup and Person schema markup you should have already placed. Together, the three form a solid identity layer that feeds knowledge graph construction and AI retrieval accuracy.

WebSite Schema and AI Search

AI Overviews and LLM-based search features use structured data as a higher-confidence input than parsed prose. The WebSite entity, especially when linked to a publisher Person or Organization, helps AI systems answer questions like "What does [site] cover?" and "Who runs [site]?" with information from your own structured declaration rather than inferred from page text.

The description property on WebSite is underused for this purpose. A well-written one-sentence description of your site's scope gives AI retrieval systems a clean, quotable summary that may appear in AI Overview answers for branded queries. It takes two minutes to add and is the kind of marginal improvement that stacks with everything else in your structured data setup.

Common WebSite Schema Mistakes

  • Placing the WebSite block on interior pages. It belongs on the homepage only — duplicating it site-wide creates conflicting entity signals and inflates crawl overhead.
  • Setting a urlTemplate that does not actually return search results. Google tests these endpoints; if the URL returns a blank or error page, the search box will not activate.
  • Omitting the publisher property. This is a missed opportunity to explicitly connect the WebSite entity to the Organization or Person entity that owns it.
  • Forgetting that Google controls whether the sitelinks search box displays. The markup signals eligibility — it does not guarantee the feature.

WebSite schema is a quick implementation that complements the entity work you do in Organization schema markup and Person schema markup. For a review of your entire structured data setup across all three identity types, schema markup consulting is where I can take a direct look and tell you what is missing or misconfigured.