Does your homepage tell AI tools your site name and how to search it?
Confirms your homepage declares the official site name and exposes your internal search to AI agents.
What this signal tests
We look at your homepage for a Schema.org WebSite block that states the canonical name of your site, its main URL, and ideally the address pattern an AI agent should use to search it. The search part is called a SearchAction and it acts like a public instruction for how to query your site programmatically.
Why it matters for your visibility in AI
When ChatGPT or Google's AI Overviews cite your site, the name they display comes from this WebSite block when it is present. Without it, AI tools fall back to whatever they can scrape from the title tag, which is often inconsistent and may include trailing slogans or campaign text that read poorly in a citation. The SearchAction has a second benefit: AI agents that are allowed to act on a user's behalf can use it to issue searches against your site and pull back fresh content rather than relying on stale cached snippets. This is increasingly relevant as agentic AI search tools roll out.
Pass criteria at a glance
| Criterion | Passes when |
|---|---|
| WebSite with name, url, valid SearchAction. |
How we test it
We fetch your homepage and look inside its JSON-LD blocks for a type of WebSite. We confirm it includes a name and a URL, and we then check whether it declares a potentialAction of type SearchAction whose target follows a URL template containing the search term placeholder. We do not actually run a search; we only verify the declaration is well-formed.
Show technical detection method
JSON-LD @type WebSite has name, url; potentialAction.@type SearchAction with target URLTemplate containing {search_term_string} and query-input.
If your site fails: how to fix it
- Add a WebSite JSON-LD block to your homepage template only. It is unnecessary on inner pages because crawlers expect it at the site root.
- Include the official site name (and an alternateName if your brand is sometimes shortened), and the canonical homepage URL with no tracking parameters.
- Add a potentialAction of type SearchAction. The target should be a URL template like https://example.com/search?q={search_term_string} pointing at your real search results page.
- Confirm your internal search endpoint actually accepts that parameter and returns relevant results. AI agents will test the link.
- Validate the block in Google's Rich Results Test, which will flag any malformed SearchAction syntax.
Quick facts
| Maturity | ESTABLISHED |
|---|---|
| Weight | medium |
| Category | Structured Data |
Primary sources
Related signals
Frequently asked questions
What if my site has no internal search?
You can still emit the WebSite block with name and url to control how your site is named in AI citations, and simply omit the SearchAction. You will lose the agentic search benefit but you will still gain consistent attribution, which is the larger of the two wins for most sites.
Will this make my site name appear differently in Google search results?
It can. Google uses the WebSite.name field as one of several inputs to determine the site name shown above search snippets. If your current site name in Google results is wrong or inconsistent, declaring it here is the most reliable way to correct it.
Does the SearchAction need to be a Google-style URL?
No. The target URL template can point to any working search endpoint on your site. The only requirement is that it includes the {search_term_string} placeholder, and that submitting a query to that URL returns a sensible result page. Custom search systems, Algolia, and ElasticSearch endpoints all work.
Run your own scan
Run a free scan and see how your site grades across all 155 AI-readiness signals.