Schema.org Markup for AI Search: The Complete Guide
Published · Updated · 10 min read
When an AI system reads your page, it sees text — but without context, it is guessing at what that text represents. Is this an article written by an expert, or boilerplate copy? Is this a step-by-step guide or a marketing paragraph? Is this answer authoritative? Schema.org markup answers those questions in machine-readable terms, directly in your page.
This guide covers the five schema types that matter most for AI citation, with working JSON-LD examples for each, and how to validate your implementation. Three of them — Organization, Article/BlogPosting and FAQPage — are the ones Aura’s 18-check scan awards points for; HowTo and WebPage are useful for AI extraction but are not scored.
Why JSON-LD Is the Right Format
Schema.org supports three encoding formats: JSON-LD, Microdata, and RDFa. For AI visibility, use JSON-LD exclusively:
- Separate from content: JSON-LD lives in a
<script>block in your page head. You do not need to restructure your HTML or add attributes to visible elements. - Easy to validate: A standalone JSON object is trivial to test and update.
- Reliably parsed: Google, Bing, and the AI crawlers that rely on them all prioritize JSON-LD. It is the most consistently supported format across pipelines.
Microdata and RDFa are still valid, but they create maintenance overhead without providing meaningful advantages for AI citability. Use JSON-LD.
1. Organization Schema
Organization schema tells AI systems who is behind the site. Without it, AI has no reliable signal about authorship — a significant trust gap. Add this to your root domain (usually your homepage), not to every page.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Your Company Name",
"url": "https://yourdomain.com",
"logo": "https://yourdomain.com/logo.png",
"description": "One-sentence description of what your organization does.",
"sameAs": [
"https://www.linkedin.com/company/your-company",
"https://x.com/yourcompany"
]
}
</script>The sameAs array links your domain to your known social profiles. This helps AI systems establish that “Aura AI Visibility” on the web and “Aura AI” on LinkedIn are the same entity — reducing ambiguity in citation.
2. Article / BlogPosting Schema
Add Article or BlogPosting schema to every piece of editorial content. This tells AI systems that the page contains citable text content, who wrote it, and when it was published. A missing publication date is a common omission that reduces citation confidence.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "Your Article Title Here",
"description": "One-sentence description of what the article covers.",
"datePublished": "2026-08-25",
"dateModified": "2026-08-25",
"author": {
"@type": "Organization",
"name": "Your Company Name",
"url": "https://yourdomain.com"
},
"publisher": {
"@type": "Organization",
"name": "Your Company Name",
"url": "https://yourdomain.com",
"logo": {
"@type": "ImageObject",
"url": "https://yourdomain.com/logo.png"
}
},
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "https://yourdomain.com/blog/your-article-slug"
}
}
</script>Use Article for editorial content. Use BlogPosting for blog entries. Use TechArticle for technical documentation. The distinction helps AI systems categorize and weight your content appropriately.
3. FAQPage Schema
FAQPage schema is the highest-impact schema type for AI citation on a per-page basis. AI systems extract Q&A pairs directly from FAQPage markup and quote them verbatim. Any page with a Frequently Asked Questions section should have this.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "What is [your topic]?",
"acceptedAnswer": {
"@type": "Answer",
"text": "A complete, self-contained answer to this question. Write it as if it will be quoted directly — because it will be. Avoid references to 'this page' or 'as mentioned above.' Include the key term in the answer text."
}
},
{
"@type": "Question",
"name": "How does [your product/service] work?",
"acceptedAnswer": {
"@type": "Answer",
"text": "A factual, citable description. Aim for 40–120 words per answer. Shorter answers get extracted cleanly; longer answers may be truncated."
}
}
]
}
</script>Write answers as standalone, quotable facts. Avoid anaphoric references (“it,” “this,” “the above”) because AI systems may extract the answer without surrounding context. Each answer should make sense in isolation.
4. HowTo Schema
HowTo schema is the right structure for procedural content: installation guides, tutorials, step-by-step walkthroughs. It breaks your content into discrete, numbered steps that AI systems can extract and present in order.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "HowTo",
"name": "How to [Do the Thing]",
"description": "A brief description of what this guide accomplishes.",
"step": [
{
"@type": "HowToStep",
"position": 1,
"name": "Step name (short label)",
"text": "Full instruction for this step. Be specific. Mention what the expected outcome is."
},
{
"@type": "HowToStep",
"position": 2,
"name": "Second step name",
"text": "Full instruction for the second step."
}
]
}
</script>Use HowTo on any page with numbered instructions. If your guide has both background explanation and step-by-step instructions, you can publish both an Article and a HowTo script block on the same page — they are compatible. Note that Aura’s scan awards no points for HowTo — it helps extraction, not your score.
5. WebPage Schema
WebPage schema identifies the canonical URL and page type for each page. It is the lightest of the five types but fills an important role: it tells AI systems the authoritative URL for the content, reducing confusion from scrapers or mirror sites. Aura’s scan does not score WebPage schema; it checks the canonical link tag instead.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "WebPage",
"@id": "https://yourdomain.com/your-page-slug",
"url": "https://yourdomain.com/your-page-slug",
"name": "Page Title",
"description": "One-sentence description.",
"isPartOf": {
"@type": "WebSite",
"@id": "https://yourdomain.com/#website",
"name": "Your Site Name",
"url": "https://yourdomain.com"
}
}
</script>How to Validate Your Schema
Two tools, used together, give you a complete picture:
- Google’s Rich Results Test — tests whether Google can parse your schema and whether your page is eligible for rich results (featured snippets, FAQ boxes). Use this for FAQPage and HowTo especially, as it shows exactly what Google will extract. Available at search.google.com/test/rich-results.
- Schema Markup Validator — validates against the full Schema.org specification. Use this to catch structural errors that Rich Results Test may not flag. Available at validator.schema.org.
Common errors to look for: missing recommended properties (Google’s rich-result guidelines recommend headline, author, and datePublished for Article; Schema.org itself has no required properties), malformed JSON (trailing commas in arrays), and mismatched types (@type: “BlogPost” is not a valid Schema.org type — it should be BlogPosting).
Schema Priority for New Implementations
If you are starting from zero, implement in this order:
- Organization — one block, homepage only, establishes authorship
- FAQPage — on every page with a Q&A or FAQ section; highest citation impact per page
- Article / BlogPosting — on every editorial or blog page
- HowTo — on any page with numbered procedural steps (not scored by Aura)
- WebPage — on all pages for canonical URL reinforcement (not scored by Aura; the scan checks the canonical link tag instead)
Frequently Asked Questions
What is JSON-LD and why is it preferred for schema markup?
JSON-LD (JavaScript Object Notation for Linked Data) encodes Schema.org markup as a script block in the page head, separate from the visible HTML. It does not require altering visible content, is easy to validate, and is reliably parsed by search engines and AI crawlers. Google and most AI systems prefer it over Microdata or RDFa.
Which schema types matter most for AI citation?
Organization, Article, and FAQPage have the highest impact — and they are the three types Aura’s scan awards points for. Start with Organization (homepage) and FAQPage (any Q&A page) if you only have time for two.
Does schema markup directly cause AI citation?
Not directly. It helps AI systems understand what your content is and who created it, which improves the probability of being selected as a citation. It is one layer of a multi-layer foundation.
How do I validate my schema markup?
Use Google’s Rich Results Test for practical validation and Schema Markup Validator (validator.schema.org) for full spec compliance.