Why structured data matters for AI search
AI answer engines use two mechanisms to retrieve content: retrieval-augmented generation (RAG), where the engine fetches and reads specific pages to answer a query, and training data, where the model draws on patterns learned during training. For RAG-based retrieval - which drives the cited answers in Perplexity and ChatGPT with browsing - structured data has a direct impact.
When an AI engine reads a page with FAQPage JSON-LD, it does not need to parse the prose to understand what questions the page answers - the question and answer are already packaged explicitly. A page without schema forces the engine to infer structure from headings, paragraph positions, and semantic patterns. That inference can fail or misattribute content. Schema removes the ambiguity.
Check which schema types are missing on your site
The free AI Search Checker tests for FAQPage, HowTo, Article and Organization schema as part of 32 AI-readiness signals. Results in 20 seconds, no login required.
Check my structured dataThe four schema types that matter for AI search
FAQPageHighest priorityWhen to use
Any page that answers questions - product pages, how-to guides, service pages, landing pages. Especially high value on pages that target "what is X" or "how to Y" queries.
What it does for AI search
Packages question-answer pairs in a format AI engines can extract directly. Each Question/acceptedAnswer pair is a citation unit - the engine reads it as: "this page definitively answers this question with this answer." FAQPage is the single schema type with the highest impact on AI citation frequency.
Template
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "What is [your topic]?",
"acceptedAnswer": {
"@type": "Answer",
"text": "[Your topic] is [clear definition in 1-2 sentences]."
}
},
{
"@type": "Question",
"name": "How does [your topic] work?",
"acceptedAnswer": {
"@type": "Answer",
"text": "[Direct explanation in 2-3 sentences]."
}
},
{
"@type": "Question",
"name": "Why does [your topic] matter?",
"acceptedAnswer": {
"@type": "Answer",
"text": "[Key benefit in 1-2 sentences]."
}
}
]
}
</script>HowToHigh priorityWhen to use
Tutorial pages, guides with numbered steps, process documentation. Pages where the content is "how to do X in N steps." Not appropriate for informational or comparison content.
What it does for AI search
Packages a step-by-step process with named steps and instructions. AI engines that answer "how to" queries can extract this structure to present your steps directly in an answer. Combine with FAQPage on the same page if the content also answers questions - the two schemas are compatible.
Template
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "HowTo",
"name": "How to [action] - step-by-step guide",
"description": "Step-by-step guide to [action] for [audience].",
"step": [
{
"@type": "HowToStep",
"position": 1,
"name": "Step 1: [Action name]",
"text": "[What to do in this step. Be specific and actionable.]"
},
{
"@type": "HowToStep",
"position": 2,
"name": "Step 2: [Action name]",
"text": "[What to do in this step. Be specific and actionable.]"
},
{
"@type": "HowToStep",
"position": 3,
"name": "Step 3: [Action name]",
"text": "[What to do in this step. Be specific and actionable.]"
}
]
}
</script>OrganizationMedium priorityWhen to use
Homepage and root layout - applies site-wide. Add once to your root layout so every page inherits it. Particularly important for brand queries: "who is [company]", "what does [company] do".
What it does for AI search
Identifies your brand to AI engines: name, URL, logo, and social profiles. AI systems use this to disambiguate your brand from others with similar names and to build a consistent understanding of what your organisation does. Without it, AI engines infer brand identity from page content alone - which is less reliable.
Template
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"@id": "https://yourdomain.com/#organization",
"name": "[Your Organisation Name]",
"url": "https://yourdomain.com",
"logo": {
"@type": "ImageObject",
"url": "https://yourdomain.com/logo.png"
},
"description": "[One sentence describing what your organisation does].",
"sameAs": [
"https://twitter.com/yourhandle",
"https://linkedin.com/company/yourcompany"
]
}
</script>ArticleMedium priorityWhen to use
Blog posts, news articles, guides with a clear author and publish date. Use og:type="article" and Article JSON-LD together. Not appropriate for tool pages, homepages, or service pages.
What it does for AI search
Marks content as authored journalism or editorial - adds author attribution, publish/update dates, and publisher identity. AI engines treat authored content differently from anonymous pages: a named author with a credential is a stronger E-E-A-T signal. Required for AI engines to cite the author's name alongside a piece.
Template
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "[Article title]",
"description": "[One sentence summary]",
"datePublished": "2026-10-01",
"dateModified": "2026-10-01",
"author": {
"@type": "Person",
"name": "[Author Name]",
"url": "https://yourdomain.com/about"
},
"publisher": {
"@type": "Organization",
"name": "[Publication Name]",
"logo": {
"@type": "ImageObject",
"url": "https://yourdomain.com/logo.png"
}
},
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "https://yourdomain.com/blog/article-slug"
}
}
</script>Implementation: where to add JSON-LD
JSON-LD goes inside a <script type="application/ld+json"> tag in the <head> section of the page. The important constraint: it must be present in the raw HTML, not injected by JavaScript after page load. AI crawlers reading raw HTML responses will not execute JavaScript - schema that only appears after JS rendering is invisible to them.
Common structured data mistakes
Frequently asked questions
What is structured data in SEO and AI search?
Structured data is machine-readable markup added to a web page that explicitly labels its content type and properties. In the context of AI search, JSON-LD structured data (FAQPage, HowTo, Organization, Article) tells AI engines exactly what the page contains - the questions it answers, the steps it describes, who wrote it, and what organisation published it - without requiring the engine to infer this from free-form prose.
Which structured data type is most important for AI search?
FAQPage is the highest-impact schema type for most pages, because it packages question-answer pairs in the exact format AI answer engines produce. HowTo has high value for tutorial and process content. Organization schema is important for brand-level citation. Article schema is important for authored content where E-E-A-T attribution matters.
Can I use multiple schema types on the same page?
Yes, and it is often correct to do so. A blog post can have both Article schema (for author and publisher attribution) and FAQPage schema (for the questions it answers) on the same page - these are compatible. HowTo and FAQPage are also compatible. Use the types that accurately describe the content - do not add schema for content that is not present on the page.
Where do I add JSON-LD structured data?
JSON-LD is added inside a <script type="application/ld+json"> tag in the <head> of the HTML page. For Next.js: add to the page's layout.tsx or use next/script. For WordPress: use the Yoast SEO plugin, Schema Pro, or add to the theme's header.php. For Webflow: add via the Page Settings > Custom Code > Head Code field. For Squarespace: add via Pages > Page Settings > Advanced > Page Header Code Injection.
How do I check if my structured data is set up correctly?
The GetInlytics AI Search Checker tests for FAQPage, HowTo, Article, and Organization schema as part of its 32-signal audit. It checks both whether the schema is present and whether it is accessible in the raw HTML (not just rendered by JavaScript). Google's Rich Results Test is also useful for validating schema syntax, though it evaluates for Google-specific rich results rather than AI search citability.
Check which schema types are missing on your site
The free AI Search Checker audits FAQPage, HowTo, Article and Organization schema alongside 28 other AI-readiness signals. It identifies which schemas are present, missing, or malformed - and returns a prioritised fix list. Free, 20 seconds, no login.
Check my structured data