FAQ Schema

No Comments
Faq schema
TL;DR: FAQ schema is FAQPage structured data that marks up question and answer pairs so machines understand them as a Q and A. Google pulled the rich result from most sites in 2023, but the markup still earns its keep by feeding AI answer engines clean, quotable pairs. Use it where you have real FAQs, not as decoration.
Type
FAQPage
Format
JSON-LD
Rich result
Mostly retired
Still useful for
AI + clarity
Abuse risk
Manual action

FAQ schema had a golden era. For a few years, adding FAQPage markup could plant a stack of expandable questions right under your listing and swallow a huge slice of the results page. Then in 2023 Google announced it would limit FAQ rich results to well known, authoritative government and health sites, which for almost everyone means the visible rich result is gone. A lot of people concluded the markup is dead. That is the wrong lesson. The rich result went away; the reasons to structure your Q and A cleanly did not.

What FAQ schema actually is

It is JSON-LD using the schema.org FAQPage type. The page contains a list of Question entries, each with an acceptedAnswer of type Answer. The rule that trips people up: the questions and answers you mark up must be visible on the page and authored by the site, not user submitted, and each question must have a single answer. It describes a page that is genuinely a set of frequently asked questions, not a product page wearing a Q and A costume.

FAQPage the page node Question name: "How do I...?" acceptedAnswer

Question name: "What if...?" acceptedAnswer One FAQPage holds many Question nodes, each with exactly one acceptedAnswer. All of it must be visible on the page.

Why it still matters after the rich result went away

Three reasons keep it worthwhile. First, AI answer engines and assistants love clean Q and A pairs. A model looking for a direct answer to a direct question finds structured FAQ content easy to lift and quote, which improves your odds of being the source it cites. Second, the discipline of writing tight, self contained answers improves the page for humans whether or not any machine reads the markup. Third, government and health sites in eligible categories can still earn the visible rich result, so for them the classic payoff remains live.

Where it goes wrong

The single biggest mistake is marking up content that is not actually an FAQ. Stuffing FAQPage onto a product page, a category page, or a checkout flow to grab attention is exactly the pattern Google's spam guidelines call out, and it can earn a manual action for structured data abuse. The markup has to reflect real, visible, on page questions and answers. The other frequent failures are mechanical: answers in the markup that do not match the visible text, HTML that breaks the JSON, or the same FAQ block duplicated across dozens of pages.

RuleDo thisNot this
Content matchMarkup mirrors the visible Q and A exactlyHidden answers only in the JSON
AuthorshipSite authored answersUser submitted Q and A (that is QAPage)
Answer countOne accepted answer per questionMultiple competing answers
Page fitReal FAQ, help, or support pagesProduct or category pages faking it
Promo contentGenuine informational answersAds, links, or coupons inside answers

How to detect problems

  • Rich Results Test and the Schema.org Validator: confirm the JSON parses and the Question and acceptedAnswer nodes are recognized.
  • Search Console: even without the rich result, the FAQ enhancement report and the manual actions report will flag markup issues and any spam penalty tied to structured data.
  • Screaming Frog or Sitebulb: crawl with structured data extraction on to find every page carrying FAQPage, spot the duplicated block problem, and catch pages where the markup exists but the visible FAQ does not.
  • Manual spot check: open the page and confirm each marked up answer is actually visible and readable, not collapsed behind something a user cannot reach or missing entirely.

How to fix it, step by step

  1. Decide whether the page is truly an FAQ. If it is a product page you dressed up, remove the markup rather than risk a manual action.
  2. Write real answers that stand on their own in two or three sentences. If a human would find them useful, a model will too.
  3. Make sure the visible text and the JSON-LD say the same thing, word for word where it counts.
  4. Use one acceptedAnswer per Question, and keep promotional junk out of the answers.
  5. Place the block only on the specific FAQ or support page it belongs to, not templated across the whole site.
  6. Validate, then recrawl to confirm it is present where intended and absent where it does not belong.
DO
  • Mark up genuine, visible, site authored FAQs
  • Keep JSON and on page text in sync
  • Write concise, self contained answers
  • Limit it to real FAQ and support pages
  • Validate after every template change
DON'T
  • Bolt FAQ schema onto product or category pages
  • Hide answers that only exist in the markup
  • Stuff links, ads, or coupons into answers
  • Duplicate the same FAQ block site wide
  • Expect the old rich result to return for most sites

What good looks like

A real FAQ or support page, valid FAQPage markup whose answers match the visible text exactly, one answer per question, no spam in the answers, and no penalty in the manual actions report. You probably will not see a rich result unless you are in an eligible category, and that is fine. The payoff now is a page that reads well for humans and hands AI systems clean, quotable pairs they can cite back to you.

FAQ

Is FAQ schema pointless now that the rich result is gone?
No. Google limited the visible rich result to eligible authoritative sites, but the markup still helps AI answer engines parse and quote your Q and A, and it enforces clearer writing. It is just no longer a guaranteed results page land grab.
Can I still get the FAQ rich result?
Per Google's 2023 change, the FAQ rich result is generally reserved for well known, authoritative government and health sites. If you are not in one of those categories, do not plan around seeing it.
FAQPage or QAPage?
Use FAQPage when your site wrote the questions and answers. Use QAPage for pages where users ask and the community answers, like a forum thread, where a question can have several competing answers.
Will bad FAQ markup get me penalized?
It can. Marking up content that is not really an FAQ, or hiding answers that only exist in the code, is structured data spam under Google's guidelines and can trigger a manual action. Keep it honest and visible.
Should the FAQ answers be visible or can they be collapsed?
They must be present in the page content. An accordion the user can expand is fine because the text is on the page. Content that only exists in the JSON-LD and never renders for a user is not.
Structured data working against you?

Misapplied FAQ markup is a common cause of structured data manual actions. An audit checks every schema type on your site, flags the risky ones, and shows you what to keep, fix, or pull.

Get an Advanced SEO Audit

Claude Vincent is a technical SEO consultant focused on crawlability, rendering, and AI-search visibility. He writes the field guides and case studies at SEO ProCheck, with a bias toward the durable, unglamorous work that decides whether search engines and AI answer engines can actually read and cite a site.

About SEO ProCheck

Technical SEO consulting and GEO strategy with 20 years of enterprise experience. Case studies, resources, and tools for search and AI visibility.

Work With Me

Technical SEO audits, GEO strategy, site migrations, and international SEO. Hourly consulting for teams who need hands-on support, not just reports.

Subscribe to our newsletter!

More from our blog