Free tool · No signup
JSON-LD schema generator
Build paste-ready Organization, WebSite and BreadcrumbList JSON-LD blocks. The single most-cited schema types across AI search engines and Google Rich Results.
Identifies the entity behind the site — name, URL, logo, contact, social links. The single most important schema for AI engines to recognise your brand.
Inputs
Your JSON-LD
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Your Brand",
"url": "https://yoursite.com",
"logo": "https://yoursite.com/logo.png",
"sameAs": [
"https://www.linkedin.com/company/your-handle",
"https://x.com/your-handle"
]
}
</script>
Drop into your <head>. Validate at Google Rich Results Test.
Why JSON-LD beats microdata in 2026
Schema.org supports three formats: microdata, RDFa, and JSON-LD. All three are technically valid. But the AI / search ecosystem has converged on JSON-LD because:
- It lives in a separate
<script>block — your HTML stays clean. - It’s easier for tooling to validate, generate and parse.
- Google explicitly recommends it.
- AI engines parse it directly without needing to walk the DOM.
If you have inline microdata, migrate to JSON-LD when you can. The Aapta GEO audit flags “no JSON-LD” even if microdata is present, because microdata is harder for AI surfaces to extract.
The three types every site should ship
Organization
One per domain, lives on the homepage. Establishes who runs the site — name, URL, logo, contact, social links. AI engines use this as the canonical entity record. Without it, queries like “who runs yoursite.com” return guesses.
WebSite
One per domain. Identifies the site as an entity (separate from the organisation behind it) and adds a SearchAction so Google can show a sitelinks search box for your brand.
BreadcrumbList
One per page. Tells AI engines where this page sits in the site hierarchy. Helps surface deeper pages in AI answers and eliminates ambiguity for nested categories.
FAQ
What is JSON-LD and why does it matter for AI?+
JSON-LD (JavaScript Object Notation for Linked Data) is the schema format AI engines and search engines parse to extract structured facts from pages. Unlike microdata, it lives in a separate <script> block and doesn't require interleaving with your HTML. AI engines preferentially cite pages that publish rich JSON-LD because the facts are unambiguous.
Which schema types matter most for AI search?+
Three to ship first: Organization (who you are), WebSite (the site itself plus a sitelinks search box), and BreadcrumbList (where this page sits in the site hierarchy). For content pages, add Article, FAQPage, HowTo or Product as appropriate. The full Aapta GEO scan flags missing types per page.
Where do I paste the JSON-LD?+
Anywhere in your <head> tag. The whole <script type="application/ld+json"> block goes in. Some templating engines have a dedicated 'head additions' field — drop it there. Most CMS / headless platforms (WordPress, Webflow, Next.js, etc.) have a code-injection setting for the <head>.
Can I have multiple JSON-LD blocks on one page?+
Yes — and you should. A typical homepage carries Organization + WebSite + BreadcrumbList. An article page carries Article + BreadcrumbList. They don't interfere with each other. Some sites combine into a single @graph object for tidiness; both work.
How do I check my schema is valid?+
Paste the URL into Google's Rich Results Test (search.google.com/test/rich-results). It tells you what structured data Google found and flags errors. Schema.org also has a validator at validator.schema.org that's stricter about edge cases.
Want a full AI-Readiness audit?
This tool fixes one signal. The full scan checks 40+ across 6 categories — bot access, structured data, content answerability, authority, technical foundations, and emerging AI standards.
Run the free AI-Readiness scan →