FAQ Schema: When to Use It (and What Changed in 2023)

What it is and how it works

FAQPage is a schema.org type that describes a page containing a list of questions and their answers. You add it as JSON-LD in the page source. Each entry is a Question with an acceptedAnswer, and the whole set lives inside one FAQPage object. The point is simple: turn a human-readable FAQ into a structured object that crawlers, validators, and language models can parse without guessing where a question ends and an answer begins.

It is meant for one specific situation: a page where you (the site) publish both the questions and the answers. It is not for forums or pages where users submit competing answers. That is a different type (QAPage), and mixing them up is a common source of validation warnings.

What changed in 2023

Until 2023, a clean FAQPage implementation could earn an expandable FAQ accordion directly in Google's search results. That visibility was the main reason most sites bothered. In August 2023 Google announced it was rolling back FAQ rich results so they would only appear for "well-known, authoritative government and health websites." For everyone else, the accordion quietly disappeared from the SERP. The markup did not become invalid, and Google did not ask anyone to remove it. The reward simply stopped arriving for the vast majority of sites.

This is the single fact that should drive your decision: if your reason for adding FAQ schema was the rich result, that reason is gone unless you run a qualifying government or health property.

The SEO and visibility angle

Here is the honest read. The classic SERP win is off the table for most sites, so do not promise a client an FAQ accordion. What remains is quieter but real. Structured Q&A is some of the most directly usable content for AI systems and answer engines, which lean on clean, machine-readable pairs when they assemble responses and citations. Well-formed FAQPage markup states plainly "this is a question, this is its answer," which removes ambiguity for any system parsing the page. That fits the broader move toward a machine-readable web, and pairs with sound content structure for AI. None of this is a guaranteed ranking lever, and you should not present it as one. It is a low-cost way to make good content legible to the systems that increasingly sit between your page and the reader. For the full argument, see our companion deep-dive on whether FAQ schema is still worth it.

Use cases: who benefits most vs least

Benefits most

Pages with a genuine, standing FAQ: product and pricing pages, service explainers, support and documentation, and help content where users really do ask the same questions. Government and health properties still get the rich result on top. Sites investing in AI visibility benefit from the clean structure regardless of the SERP.

Benefits least

Pages with no real questions, where an FAQ would be invented purely to carry markup. Thin pages, doorway-style content, and anywhere the "answers" are actually promotional copy. If you would not write the FAQ for a human reader, the schema adds nothing and creates risk.

Implementation

Add the markup as JSON-LD in the page head or body. Every question and answer in the markup must also be visible to the user on the page. Here is a complete, valid example.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "Does FAQ schema still show a rich result in Google?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "For most sites, no. Since 2023 Google reserves the FAQ rich result for well-known authoritative government and health sites."
      }
    },
    {
      "@type": "Question",
      "name": "Should I remove FAQ schema I already have?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "No. Valid FAQPage markup is still useful for machine readability and remains eligible if Google changes the policy."
      }
    }
  ]
}
</script>

Keep answers plain and factual. No links, buttons, or promotional language inside the text field, and no HTML beyond the limited tags the spec allows.

How to validate

Run the markup through Google's Rich Results Test and the Schema.org Validator. The Rich Results Test confirms the syntax is parseable, though for most sites it will note that the FAQ enhancement is limited to eligible properties. The Schema.org Validator checks the structure against the vocabulary itself. Treat a clean parse, not a rendered preview, as success here. Spot-check that every question and answer in the JSON-LD matches what a visitor actually sees on the page.

Common mistakes

• Marking up content that is not on the page. If a question or answer does not appear visibly to the user, it should not be in the markup. This is the most frequent and most penalizable error.
• Promotional answers. Coupon codes, calls to action, and sales copy in the text field violate the guidelines and can trigger warnings.
• Using FAQPage for user-generated Q&A. If visitors submit the answers, you want QAPage, not FAQPage.
• Inventing FAQs to carry schema. A fake FAQ block helps no one and dilutes the page.
• Letting markup drift. When you edit the visible answer, edit the JSON-LD to match.

FAQ