SEO

FAQPage schema

Q: What is FAQPage schema?

FAQPage schema is a schema.org type for question and answer pages. It contains mainEntity as an array of Question objects with acceptedAnswer. The result: extended rich snippets in the Google SERP with expandable questions.

Q: Does Google always show FAQ rich results?

No longer, for any site. Google stopped showing FAQ rich results on 7 May 2026, after restricting them to government and health sites in August 2023. FAQPage schema is still worthwhile nonetheless: it is used by LLMs (ChatGPT, Perplexity) for citation extraction.

Q: How do you implement FAQPage schema?

As JSON-LD in the HTML head with a mainEntity array. One Question object per question, with name (the question) and acceptedAnswer (the answer with text). Important: the visible FAQ content and the schema must match.

Q: How many FAQs make sense per page?

5–10 per page is the sweet spot. Too few fail to use the schema's potential, too many appear artificial. Focus on questions that are genuinely asked frequently — not generic ones, but those relating to the specific content topic.

Also known as: FAQ schema, FAQPage JSON-LD, FAQ markup

The FAQPage schema is a Schema.org type that explicitly signals to a web page: ”Here are question/answer pairs.” Each entry consists of a Question object (with name) and a nested acceptedAnswer object (with text). The markup is usually placed as JSON-LD in the <head> or directly alongside the FAQ section. Google has not shown FAQPage markup as a rich snippet at all since 7 May 2026, but the structured data remains essential for AI citations: ChatGPT, Perplexity, Gemini and Google AI Overviews parse Q&A structures far more reliably when they are explicitly typed.

Minimal example

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "What does the tool cost?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Pay-per-use from 5 cents per action."
      }
    }
  ]
}

One further entry object in mainEntity per question. Answers allow plain text with limited HTML (paragraphs, lists, links).

When is a FAQPage appropriate?

A standalone FAQ page (e.g. /faq) with ≥3 entries.
An FAQ section at the end of product or service pages.
Knowledge base articles in question/answer format.

NOT suitable: marketing CTAs (”Ready to get started?”), rhetorical questions in body copy, forum threads with user answers (QAPage is intended for those).

Why Google has shown no FAQ snippets since May 2026

This happened in two stages. In August 2023 Google restricted the display to government and health sites. On 7 May 2026 it disappeared entirely: since then FAQ results no longer appear in Google Search for any site. In June 2026 the search appearance filter, the Search Console report and Rich Results Test support were removed, and in August 2026 support in the Search Console API followed.

The reason to add it anyway has shifted. Where Google once showed an expandable element, language models now take the answer and name the source. A passage marked up as Question with acceptedAnswer does not have to be extracted from body copy first — it is already a finished answer.

Where the markup fails

Invisible content — every marked-up question and its answer must also be visible to visitors. Markup describing something nobody can see violates Google’s structured data guidelines.
Visible content without markup — the reverse case, and the more common one: the page shows questions and answers, the markup is missing. Rankmio calls this content drift.
The wrong type — FAQPage applies when the site owner answers. If a single question is the focus and users answer it, QAPage is the right type.
Answers that are not answers — “Get in touch for more information” answers nothing. Filling the markup with promotional copy gives language models nothing they could quote.

How does Rankmio generate the FAQ page?

Since 24 June 2026 the Content Studio editor has included a schema helper that recognises FAQ drift (Q&A structures in the content without FAQPage schema) and offers a one-click generator: ”FAQ schema (0 Cr)” on the full-check item builds a valid FAQPage JSON-LD heuristically from the recognised pairs — no AI call, no risk of hallucination, no credit consumption.

Example from practice

Example: A SaaS landing page had an FAQ section with 5 genuine entries but no FAQPage schema. In the GEO audit, ChatGPT citations stood at 0%. After the markup was added (one click in the editor), the citation rate rose to 14% over the following 4 weeks — the AI systems could now cite the answers directly as a source without having to guess the Q&A structure.

Frequently asked questions

What is FAQPage schema?

FAQPage schema is a schema.org type for question and answer pages. It contains mainEntity as an array of Question objects with acceptedAnswer. The result: extended rich snippets in the Google SERP with expandable questions.

Does Google always show FAQ rich results?

No longer, for any site. Google stopped showing FAQ rich results on 7 May 2026, after restricting them to government and health sites in August 2023. FAQPage schema is still worthwhile nonetheless: it is used by LLMs (ChatGPT, Perplexity) for citation extraction.

How do you implement FAQPage schema?

As JSON-LD in the HTML head with a mainEntity array. One Question object per question, with name (the question) and acceptedAnswer (the answer with text). Important: the visible FAQ content and the schema must match.

How many FAQs make sense per page?

5–10 per page is the sweet spot. Too few fail to use the schema's potential, too many appear artificial. Focus on questions that are genuinely asked frequently — not generic ones, but those relating to the specific content topic.

Used in Rankmio for

FAQ schema generator in Content Studio

Go to the feature →

Last updated: 2026-09-01 · Browse all glossary entries