How FAQPage schema works
FAQPage marks up a page that asks questions and answers them itself. Each pair becomes a Question with an accepted answer attached, so a machine reading the page can separate the question text from the response instead of seeing one continuous block of prose.
It is distinct from QAPage, which describes a very different kind of page: one where a user posts a question and other people submit competing answers, as on a forum or a community support site. Using the wrong one of the two describes the page inaccurately, so check which situation you are actually in before choosing.
The markup only ever labels what is already published. The questions and answers must be visible to the visitor on the page itself, in full — not hidden behind a tab that never loads, and not written for the markup alone.
Why FAQPage schema matters — and matters less than it did
For several years this was the easiest rich result to win. Marking up an FAQ block could pull an expandable list of questions into the search result, taking up more of the screen and pushing competitors down. That is largely over. Google now restricts FAQ rich results to well-known authoritative government and health sites, so an ordinary business page will not get the expandable listing however good the markup is.
What remains is quieter but real. The markup still tells search engines and AI answer systems exactly which sentence answers which question, which is useful when a system is looking for a short, attributable response. It costs almost nothing to publish correctly, and it should be sold to nobody as a way to take over the search results page.
Common mistakes with FAQPage schema
The first is still expecting the rich result, and building content around a feature that will not appear. Pages get stuffed with invented questions nobody asks, in the hope of a listing that is no longer available, and the page reads worse for real visitors as a result.
The second is marking up questions that are not on the page, or answers that differ from the visible text. The third is applying FAQPage to a whole page whose main purpose is something else entirely — a product page or a checkout flow with a few help notes at the bottom is not a FAQ page, and describing it as one misrepresents what the page is.
How to act on it
Write the FAQ for the reader first. If the questions are ones customers genuinely ask — about price ranges, timelines, what happens after they enquire — the section earns its place whether or not a search feature ever picks it up, and it gives your chance at a featured snippet something concrete to work with.
Then mark up what you published, keeping the schema text identical to the visible text, and validate with the Rich Results Test. Do not remove existing FAQ markup out of disappointment; it is still an accurate description of the page. Just set the expectation honestly with whoever signed off the work, and put the effort you would have spent chasing the rich result into the content itself.