What is FAQPage schema?

FAQPage schema is a Schema.org structured data type that marks up a page containing a list of questions and answers. It tells search engines and AI models exactly which questions your page answers and what the precise answers are.

How do I add FAQPage schema to my website?

Add a script tag with type application/ld+json to your page's HTML head, containing a JSON object with @type set to FAQPage and a mainEntity array of Question and Answer objects. Each question needs a name and an acceptedAnswer with the answer text.

How many FAQs should I include in FAQPage schema?

Include 5 to 10 questions that directly match what users search for. Prioritize questions with clear, concise answers under 300 words. More is not always better — focus on questions with genuine search demand.

Can I use FAQPage schema on blog posts?

Yes. You can add FAQPage schema to any page that contains questions and answers, including blog posts, landing pages, and product pages. It does not need to be a dedicated FAQ page.

Why does FAQPage schema help AI citation?

FAQPage schema removes all inference work from AI engines. Each Question and Answer pair is explicitly labeled with a clear semantic boundary, mapping directly to the question-and-answer format AI was trained to produce. AI engines can lift FAQ answers directly and cite the source with high confidence.

What makes a good FAQ answer for AI citation?

A good FAQ answer for AI citation is complete on its own (no 'read more' needed), 40–160 words, free of promotional language, written in plain text with no HTML tags, and phrased as if answering a voice assistant query.

What is the difference between FAQPage, HowTo, and Speakable schema?

FAQPage marks up Q&A pairs for AI extraction. HowTo marks up numbered step-by-step instructions. Speakable identifies a 280-350 character passage for voice assistant responses. They serve different query types and can all be used on the same page.

FAQPage Schema: How to Get Your FAQs Cited by AI Answer Engines

FAQPage schema is the fastest schema type to implement and the highest-impact one for AI citation. It structures your questions and answers into a machine-readable format that AI engines can extract and present verbatim — no inference required. Answer engines treat FAQPage markup as pre-digested, citation-ready content.

Why AI engines love FAQPage schema

AI answer engines operate under computing constraints. They prefer content that requires minimal inference to extract a fact. FAQPage schema removes all ambiguity: each Question and Answer pair is explicitly labeled, has a clear semantic boundary, and maps directly to the question-and-answer format that AI was trained to produce.

A page with FAQPage schema is essentially pre-formatted as AI output. The engine can lift it directly, cite the source, and move on. That efficiency is why FAQPage-tagged pages get cited disproportionately often. To understand the bigger picture of how AI citation works, see What Is Answer Engine Optimization (AEO)? A Plain-English Guide.

The same signal that makes FAQPage valuable for AI is what made it valuable for Google's featured snippets: a machine-readable declaration of exactly what question a piece of content answers. Google reduced the visual display of FAQ rich results in standard search in 2023, but the underlying extraction mechanism — which powers AI search — still relies heavily on this schema type.

The science behind why AI engines love FAQPage

AI search engines operate under a constraint called the answer budget — a limit on how much compute they can spend retrieving and verifying each answer. When you ask ChatGPT or Perplexity a question, the model doesn't read every word of every web page. It processes structured signals first, then falls back to raw text when structure is absent.

FAQPage schema compresses the answer budget cost to near-zero. Instead of:

Parse the raw HTML of 50 pages
Identify paragraphs that might contain the answer
Score them by relevance and confidence
Select the best passage

The AI engine does:

Read FAQPage.mainEntity[0].name — this is the question
Read FAQPage.mainEntity[0].acceptedAnswer.text — this is the answer
Cite it

That's it. The explicit structure eliminates steps 1–3 entirely. Pages with FAQPage schema get processed first, cost less compute to extract, and produce higher-confidence citations.

What FAQPage schema looks like

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "What is Answer Engine Optimization?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Answer Engine Optimization (AEO) is the practice of structuring content so AI-powered engines like ChatGPT and Perplexity select it as their definitive answer."
      }
    },
    {
      "@type": "Question",
      "name": "How do I add FAQ schema to my website?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Add a <script type='application/ld+json'> block in your page's <head> containing the FAQPage JSON-LD. Validate it with Google's Rich Results Test before deploying."
      }
    }
  ]
}

For the full JSON-LD implementation syntax and how to inject it correctly, refer to JSON-LD for Bloggers: How to Add Structured Data to Blog Posts.

The hard-coded HTML rule

AI crawlers — GPTBot (ChatGPT), PerplexityBot (Perplexity), ClaudeBot (Anthropic) — often do not execute JavaScript. If your FAQPage schema is injected by a script after page load, these bots will never see it.

Rule: FAQPage JSON-LD must be hard-coded in static HTML inside <head>.

Verify this by opening your page in a browser and pressing Ctrl+U (Windows) or Cmd+U (Mac) to view the raw page source. If the <script type="application/ld+json"> block appears in that view, AI crawlers can read it. If it only appears in browser DevTools → Elements panel after a delay, it is JavaScript-injected and invisible to most AI bots.

Every major CMS has a field for static head injection:

WordPress + Rank Math / Yoast SEO — generates FAQ schema automatically from FAQ blocks
Webflow — "Custom Code" → Head section in Page Settings
Framer — Page Settings → Custom Code → <head>
Ghost — Settings → Code injection → Site header
Next.js — Server component with dangerouslySetInnerHTML (renders to static HTML)

How to implement FAQPage schema: step by step

Step 1 — Identify your Q&A pairs

Look for existing FAQ sections on your page. If none exist, create 5–10 questions your audience actually asks about your topic. Answer each in 1–3 sentences with a clear, complete answer.

The best FAQ questions are:

Phrased exactly as a user would type or speak them
Specific enough to have a clear, definitive answer
Covering different aspects of the topic (not all asking the same thing with different wording)
Matched to real search queries (use Google Autocomplete, Reddit, and Quora to find actual phrasing)

Step 2 — Write answers that get cited

The 8 rules for AI-citation-ready FAQ answers:

Complete on their own. The answer must satisfy the question without the reader needing to visit your page. If the AI can't reproduce the full answer, it won't cite you.
40–160 words. Short enough to be read aloud or displayed inline; long enough to be substantive.
Plain text only. Strip <em>, <strong>, <a>, and any other HTML from the text field. Use plain prose.
No promotional language. Answers that say "Our award-winning tool is the best solution..." get skipped. Factual, neutral tone wins.
Present tense where possible. "AEO targets AI answer engines" ages better than "In 2026, AEO is used to target AI answer engines."
Include the key term. The answer text should include the core concept you want AI to associate with your page.
Address the whole question. If the question has multiple parts ("What is X and how do I use it?"), address both parts in the answer.
Write as if answering a voice assistant. Voice assistants read FAQ answers aloud. Write answers that sound natural when spoken.

Step 3 — Write the JSON-LD

Use the template below. Replace each name with the full question text including the question mark. Replace each text in acceptedAnswer with the verbatim answer in plain text.

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "What is [your topic]?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Your complete, plain-text answer here. No HTML tags."
      }
    },
    {
      "@type": "Question",
      "name": "How do I [action]?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Your complete, plain-text answer here."
      }
    }
  ]
}

Step 4 — Place it in `<head>`

<head>
  <script type="application/ld+json">
  {
    "@context": "https://schema.org",
    "@type": "FAQPage",
    "mainEntity": [...]
  }
  </script>
</head>

Step 5 — Validate with Google Rich Results Test

Visit search.google.com/test/rich-results, enter your page URL, and confirm FAQPage shows as detected. Fix any errors before submitting the URL for re-indexing in Google Search Console.

FAQPage vs HowTo vs Speakable: when to use which

Schema type	Best for	Query type
`FAQPage`	Q&A content, "what is", "does X do Y"	Informational queries
`HowTo`	Step-by-step instructions	How-to queries
`SpeakableSpecification`	Short 280–350 char passage	Voice queries

These are complementary, not competing. A how-to guide should have HowTo schema for the steps and FAQPage schema for the Q&A section at the bottom and SpeakableSpecification pointing at the opening summary paragraph. Stack them in a single @graph block.

Advanced FAQPage techniques

Multi-topic FAQ pages

If your page covers multiple distinct topics, group your FAQ entries by topic using a clear heading structure in your HTML. The FAQPage schema treats the entire mainEntity array as a flat list — but AI engines use the surrounding HTML context to understand how questions cluster.

Keep related questions together both in the JSON-LD array and in the visible HTML. An AI engine reading a page with a coherent topical structure is more likely to cite it for multiple queries.

Combine with BlogPosting using @graph

The most effective pattern for articles: nest FAQPage inside a @graph block alongside BlogPosting and BreadcrumbList.

{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "BlogPosting",
      "@id": "https://yoursite.com/articles/slug#article",
      "headline": "..."
    },
    {
      "@type": "FAQPage",
      "@id": "https://yoursite.com/articles/slug#faq",
      "mainEntity": [...]
    },
    {
      "@type": "BreadcrumbList",
      "itemListElement": [...]
    }
  ]
}

Update answers regularly

FAQ answers with specific years, version numbers, or statistics go stale. An answer that says "In 2024, 45% of sites use schema" is less trustworthy than "As of the latest W3Techs data, approximately 45% of sites use schema." Review your FAQ answers quarterly and update any time-anchored claims.

FAQPage audit checklist

Before submitting for indexing, verify all 10:

✅ Schema is in static HTML (confirmed via View Source, not DevTools)
✅ No JSON syntax errors (validate at jsonlint.com)
✅ No HTML tags inside acceptedAnswer.text fields
✅ At least 5 questions included
✅ Each answer is 40–160 words
✅ Answers are in plain, factual language (no promotional phrasing)
✅ Question name fields include the question mark
✅ No duplicate answer text copied from other pages
✅ Google Rich Results Test shows FAQPage as detected
✅ URL submitted for re-indexing in Search Console after deployment

Common questions about FAQPage schema

How many questions should a FAQPage schema have?

Google recommends at least 2 questions. For AI citation coverage, 5–10 questions is the sweet spot — enough variety to match different query phrasings, without diluting the per-answer quality. More than 10 is rarely better: AI engines tend to sample from the first 5–7 entries.

Does FAQPage schema still work in 2026?

Yes. While Google reduced rich-result display of FAQ data in 2023, the schema still provides significant AEO benefit — AI engines index and cite it independently of Google's visual presentation choices. ChatGPT, Perplexity, and Claude all use FAQPage markup to identify citation-ready Q&A content.

Can I have multiple schema types on one page?

Yes. Use the @graph pattern to combine FAQPage with Article, BreadcrumbList, and SpeakableSpecification on the same page. Each schema type provides a different extraction signal.

What is the difference between FAQPage for Google and FAQPage for AI?

For Google's rich results, FAQPage previously generated expandable Q&A directly in the search result. That visual feature was reduced in 2023. For AI engines, FAQPage works differently — it's used during training and crawling to identify content that maps directly to question-answer pairs. The benefit is AI citation, not visual SERP enhancement. Both audiences benefit from the same well-written, correctly implemented FAQPage schema.