Structured data (schema) for AI: what to add

Structured data (schema, usually in JSON-LD format) tells Google and AI engines what each page is and who stands behind it. It cuts ambiguity and improves your eligibility for rich results and for being cited in AI answers.

For AEON42, structured data isn't a stand-alone checkbox. It's one of the technical signals we evaluate alongside everything else, because AEON42 unifies SEO, GEO, and AEO on top of your real Search Console data. The same engine that reads your clicks, impressions, and positions also scans your schema and tells you what's missing for your industry.

What to add, by priority

1. Organization — the entity behind the site: name, logo, contact, social profiles. This is the foundation of your identity for AI. When ChatGPT, Perplexity, or Gemini try to figure out who you are, a clean Organization block removes the guesswork.
2. WebSite — your site as an entity; optionally with SearchAction if you have an internal search box.
3. Article — on every article or guide: headline, description, publish date, author. This is what makes a piece eligible to be quoted and attributed.
4. FAQPage — on pages with frequently asked questions; it helps you surface in direct answers and AI Overviews.
5. BreadcrumbList — the navigation path; it gives hierarchical context so engines understand where a page sits in your site.

For a SaaS, SoftwareApplication additionally describes your product and its plans. For products, services, events, recipes, or local businesses, there are dedicated types — match the schema to what the page actually is rather than over-marking everything.

The order above is a priority, not a wish list. If you only have time for two, ship Organization and Article: one anchors your brand as an entity, the other makes your content citable.

How to implement it

• Use JSON-LD (Google's recommended format): a <script type="application/ld+json"> block in the HTML. It's the easiest to maintain because the markup lives separately from your visible content.
• Render it server-side so any crawler can read it without executing JavaScript. Many AI crawlers don't run JS, so client-only schema is invisible to them.
• Link entities to each other with @id. For example, an Article points to your Organization as its publisher, and your WebSite references the same Organization. A connected graph is far more legible to an engine than five disconnected blocks.
• Keep schema consistent with what's on the page. Markup that claims a rating, price, or author the page doesn't actually show is a quality violation and can get you ignored or penalized.
• Validate with Google's Rich Results Test and with validator.schema.org before you ship. Both catch syntax errors and missing required fields in seconds.

Here is a minimal, connected example tying an Organization to an Article:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Organization",
      "@id": "https://example.com/#org",
      "name": "Example Co",
      "url": "https://example.com",
      "logo": "https://example.com/logo.png"
    },
    {
      "@type": "Article",
      "headline": "How structured data helps AI cite you",
      "datePublished": "2026-06-19",
      "author": { "@type": "Person", "name": "Jane Doe" },
      "publisher": { "@id": "https://example.com/#org" }
    }
  ]
}
</script>

Notice the @id reuse: the Article's publisher references the Organization block instead of repeating it. That's the pattern an engine follows to resolve your entity.

Does schema replace good content?

No — and this is the common misread. Structured data describes content; it doesn't create authority. An engine still has to find a clear answer on the page worth quoting. Schema makes that answer easier to identify and attribute, but a thin page with perfect markup still won't get cited. Pair schema with citable passages, real expertise (E-E-A-T), and pages that AI crawlers can actually reach. See how to allow AI crawlers for the access side of that equation.

How to know what you're missing

AEON42 scans your homepage schema and compares it against what's expected for your industry, then tells you exactly which types you're missing — without naming things in jargon you'd have to decode. Because that scan lives next to your Search Console performance, you can see which underperforming pages would benefit most from better markup, and prioritize accordingly.

That's the whole point of unifying SEO, GEO, and AEO on one dataset: a schema gap stops being an isolated technical note and becomes a ranked, prioritized fix tied to real traffic.

See pricing · or read how to appear in Google AI Overviews.