Guides

FAQ Schema: What It Is and What It Actually Does Now

FAQ schema is structured data, written as JSON-LD, that labels a set of questions and their matching answers on a page so search engines and AI systems read them as a defined list of question and answer pairs instead of guessing from your raw HTML. You place it on pages that already contain genuine questions and answers, and it marks precisely which text is the question and which text is the acceptedAnswer tied to it.

Here is the part most guides leave out. Since 2023, Google stopped showing the expandable FAQ rich result for ordinary business sites and reserved that treatment for recognized government and health authority domains. If you are adding faq schema to win back the dropdown accordion under your listing, that is gone for almost everyone, and no amount of valid code changes it. The markup still earns its place, but for different reasons than the ones people remember from a few years ago.

What FAQ schema actually is

The markup draws on the FAQPage type from the shared Schema.org vocabulary. Inside the FAQPage object sits a list, normally named mainEntity, holding one or more Question items. Each Question carries a name, which is the question text exactly as a reader would ask it, and an acceptedAnswer of type Answer, which holds the response. That nesting is the entire concept. You are handing a crawler a clean, pre-separated structure, so it never has to infer from heading tags and paragraph spacing which sentence is a question and which block answers it.

FAQPage: the wrapper that declares the page documents a list of frequently asked questions.

Question: one entry, with its name set to the exact question a reader would type or speak.

acceptedAnswer: the answer bound to that question, written as plain text or the limited HTML the vocabulary permits.

You can write this block by hand or generate it from your content management system. Either way the output is a small piece of JSON that lives in the page source, usually inside a script tag. It is invisible to visitors and read only by machines, which is why it has to match what the page shows rather than inventing anything new.

What it does now, and what it stopped doing

Treat these as two separate questions, because the honest answer differs for each one.

What it stopped doing

It no longer generates the expandable FAQ rich result in standard search results for most sites. Adding valid markup will not bring back that accordion of collapsible answers, and tweaking the code endlessly will not either, because the limitation is a policy decision, not a technical bug on your end. Chasing that old rich result is the single most common reason owners feel let down by this markup.

What it still does

The quieter benefits are real, and they are the reason to keep using it with intent rather than abandon it.

Gives search engines an unambiguous read of your question and answer content, which removes guesswork about how the page is structured.

Gives AI answer engines clean, labeled question and answer pairs to draw from when they assemble a response, which matters as more discovery moves into generated answers.

Forces you to write real questions in the words your buyers actually use, which tends to sharpen the page itself.

Stays valid and future friendly, so if eligibility ever widens again, the structure is already in place.

That second point deserves weight. AI answer engines build responses by pulling and recombining passages. A page that hands them labeled question and answer pairs is easier to quote accurately than one that buries the same information in long prose, which is exactly why this markup matters more in an AI search context than it did when it was only about a search accordion.

When to use FAQ schema, and when to skip it

Use it on pages that genuinely answer recurring questions: a service page that handles real objections, a product page with honest buyer questions, a support article that resolves a common problem. Skip it the moment you would have to invent questions just to justify the code. Fabricated FAQs make a page longer and weaker, and they cut against the guidelines even when the markup validates cleanly. The test is simple: would a human reader thank you for this question block, or is it there only for a machine?

Good fit: service pages, pricing explainers, how-it-works guides, and support docs with real recurring questions.

Poor fit: thin pages, promotional blocks disguised as questions, or the same FAQ copied unchanged across many pages.

How to implement it correctly

The rules are strict, and following them exactly matters, because sloppy or dishonest markup is worse than none. A single structured data block that lies about your page can cost you trust with a crawler that would otherwise read you cleanly.

Write the JSON-LD block and place it in the page source. One FAQPage object per page is the clean approach.

Every question and answer in the markup must match content that is visible to a human on that same page. Hidden-only markup is a guideline violation.

Keep answers in plain text or the limited HTML the vocabulary allows. Do not stuff links, promotions, or keywords into the answer field.

Validate before you ship. Run the markup through Google's Rich Results Test and the Schema.org validator to catch type and nesting errors.

Common mistakes to avoid

Most failures come down to a handful of repeat offenders.

Marking up questions that are not shown anywhere on the page.

Duplicating the same FAQ block across dozens of pages to game coverage.

Writing promotional copy in the answer field instead of a real answer.

Expecting a rich result that the current policy no longer grants to most sites.

Leaving the markup out of sync with the visible text after a later content edit.

Where FAQ schema fits in a wider plan

FAQ schema is one small, honest piece of a structured data strategy, not a growth lever on its own. It works alongside other markup types and, more importantly, alongside content that actually answers the questions people ask. Ninetyfifth Marketing bakes this kind of structure into the sites it builds as part of ongoing SEO and AI search work, rather than bolting it on afterward. If you want the method behind that approach, the related guides below go deeper, and the service pages explain how the work is scoped.

Questions

Does FAQ schema still get a rich result in Google?

For most websites, no. Since 2023 Google limits the expandable FAQ rich result to recognized government and health authority sites. Adding valid markup to a typical business site will not produce the accordion, though the markup is still read and still useful for other reasons.

Is FAQ schema worth adding if there is no rich result?

Often yes, when the page has genuine questions. The markup gives search engines and AI answer engines clean, labeled question and answer pairs, which makes your content easier to parse and quote accurately. It is not worth adding to a page where you would have to invent questions to use it.

Can I put FAQ schema on every page of my site?

You should not. Only mark up pages that actually show real frequently asked questions to visitors. Duplicating the same FAQ block across many pages, or marking up questions that are not visible, works against the guidelines and can erode trust with crawlers.

What is the difference between FAQ schema and regular schema markup?

FAQ schema is one specific type within the broader schema markup vocabulary. Schema markup is the general system for labeling any content, while FAQPage markup is the narrow type for a list of questions and their answers. You choose the type that matches the content on the page.

Do the answers in the markup have to match the visible page?

Yes. Every question and answer in the markup must match content a visitor can see on that same page. Hidden-only markup, or answers that differ from the visible text, is a guideline violation and can get the structured data ignored.