TacticUnprovenIn the free check · info only

FAQPage / HowTo structured data

FAQ/HowTo schema is hygiene, not a citation boost. Ahrefs' matched study of 1,885 pages that added schema found no meaningful AI citation uplift, and Toronto found FAQ blocks null within a site. Bing using schema so Copilot can understand content is not the same as being cited.

Little/no measured citation lift, only thin support, or a mechanism without outcome data.

Little or no citation lift, or only thin support. Hygiene if the check still reports it.

How we grade →

How to fix

Nothing needs fixing. No study we trust shows FAQ or how-to markup gets you quoted by AI agents, and Google stopped showing HowTo rich results in 2023 and FAQ rich results in 2026. The free check reports it but doesn't grade it.

If you keep it:

  1. Mark up only questions and answers that are on the page, word for word (see "Keep FAQ markup matching the page" below).

  2. Use one FAQPage block per page, in JSON-LD:

    <script type="application/ld+json">
    {"@context":"https://schema.org","@type":"FAQPage","mainEntity":[
      {"@type":"Question","name":"How much does Acme Payroll cost?",
       "acceptedAnswer":{"@type":"Answer","text":"Plans start at £29 a month for up to 5 people."}}
    ]}
    </script>
    
  3. Test it with Google's Rich Results Test or the Schema Markup Validator.

How to check it worked: run the free check again. The "FAQ or how-to markup" row reads "You have FAQ or how-to markup. Fine to keep, but there's no evidence it gets you quoted."

Keep FAQ markup matching the page

Make sure every question in your FAQ markup is also on the page. Google says: "Don't mark up content that is not visible to readers of the page."

  1. List the questions in your FAQPage JSON-LD: View Source and search for "Question".
  2. Check that each question and answer appears on the page in the same words.
  3. Add any missing ones to the page, or remove them from the markup.
  4. Build the markup from the same source as the visible FAQ if you can, so they can't drift apart. Otherwise, update both whenever you edit the FAQ. Folded accordions are fine if the text is in the HTML.

How to check it worked: run the free check again. The "Your FAQ markup matches the page" row reads "All {n} FAQ questions in your markup are on the page."

What the free check looks for

  • Parse JSON-LD/microdata for FAQPage/HowTo; validate syntax; report presence as minor. Check visible on-page Q&A content (see answer-first-question-headings) higher than the markup itself.

Caveats

  • — Correlational studies dominate this field. AI citations churn: a small 15-prompt preprint (arXiv 2609.22655; the author sells AI-visibility tooling) saw each engine's cited URL set change by about 67% on average from one day to the next, with under 1% URL overlap across engines. Toronto (2610.15033) found only about 40% of the within-site pattern of ChatGPT citations still held two weeks later, and 14% after ten weeks. Re-verify yearly.
  • — Hygiene only — reported as info, not graded in the free check. No study shows FAQ or HowTo markup raises AI citations.

Studies and docs

2023

Google limited FAQ rich results to authoritative gov/health sites and deprecated HowTo rich results.

2025

Three single-page test sites (good, poor, no schema): only the well-marked-up page appeared in an AI Overview, but the no-schema page was never indexed, so the comparison is confounded. Authors call it promising but inconclusive.

2025

Of ~9,000 citations from 2,000+ prompts on ChatGPT, AI Overviews and Perplexity, only 1.8% of cited pages used FAQPage schema and 6.9% Product. Descriptive (no comparison with uncited pages).

2025

Visibility distributions in Perplexity, Gemini and OpenAI looked nearly identical across schema-coverage buckets: no consistent relationship between schema and AI visibility. Vendor study.

2025

Vendor statement that Bing uses schema to help Copilot/LLMs understand content. Understanding is not citation lift; no outcome data.

2026

Matched before/after (difference-in-differences) study of 1,885 already-cited pages that added schema: no meaningful AI citation uplift (AI Overviews −4.6%, significant; AI Mode +2.4% and ChatGPT +2.2%, not significant). Observational, not a randomized test. Hygiene, not a citation boost. Do not claim schema → 3× citations.

← All tactics · How we grade

See what AI agents can find on your site

Our free AI visibility check shows whether AI agents can reach, read and quote your site, with each fix graded by evidence. There's no overall score, and no one can promise citations.

Free. No password.