What is FAQPage schema, and should you use it?
Markup that labels a genuine FAQ section so machines can read the questions and answers as questions and answers, rather than as an undifferentiated block of text. Useful when you have a real FAQ, and actively risky when you pretend you do.
What it actually is
A page can look obviously like an FAQ to a person — bold question, answer underneath, repeat — and be completely ambiguous to a machine, which sees only a sequence of headings and paragraphs. FAQPage markup makes the relationship explicit: this text is a question, that text is its answer, and the same business wrote both.
Why it matters
Question-and-answer content is the single most quotable shape on the web for AI assistants, because it matches what a user asked. When your answer is already labelled as the answer to a specific question, an assistant can lift it with confidence and attribute it to you. Unlabelled, that same answer competes with every other paragraph on the page for the assistant's attention.
What good looks like
Markup that appears on pages with a real FAQ — several genuine questions, each with your own substantive answer — and matches what a visitor can actually see on the page. The questions in the markup should be the questions on the page, and the answers should be the answers. It belongs on pages that have an FAQ, not on every page of the site.
Being honest about it
Google narrowed the rich result this markup earns, so the visible search payoff is much smaller than it was a few years ago and anyone selling it on rich snippets alone is behind. The reason it still matters is AI readability, not decoration in the results page. And it genuinely does not apply to most pages — a glossary, a service page or a blog post does not need it, and we do not treat its absence there as a problem.
Where most sites go wrong
The dominant failure is markup describing an FAQ the page does not have, usually inherited from a template or an SEO plugin applied site-wide. That is a mismatch between what you claim and what you show, which is the category of problem search engines treat most harshly. A near relation is marking up questions the site answers elsewhere, or not at all, leaving an assistant holding a question with an empty answer attached to your name.