FAQ schema is one of the most misunderstood types of structured data. Some site owners slap it on every single page, hoping to win rich results everywhere. Others use it only on a single dedicated FAQ page and wonder why it never seems to do much. Neither approach is quite right.
The real answer depends on where questions and answers actually exist on your site, what Google and AI engines reward, and what you're actually trying to achieve. Let's work through it properly.
What FAQ Schema Actually Does
FAQ schema uses the FAQPage type from Schema.org to mark up a page that contains a list of questions and their corresponding answers. When implemented correctly as JSON-LD, it signals to search engines and AI systems that specific content on a page is structured as a question-and-answer format.
In Google Search, this has historically produced rich results showing expandable questions directly in the SERP. Google has scaled back how often these appear for commercial and product pages, but the markup still carries weight for AI search engines like ChatGPT, Perplexity, and Gemini. These systems actively look for well-structured Q&A content when forming responses to user queries. If your page answers a question clearly and that answer is marked up with FAQ schema, you have a much better chance of being cited.
That distinction matters. FAQ schema is not purely a Google rich-results play anymore. Its value has shifted considerably towards AI readability.
The Core Rule: Schema Should Reflect What Is Actually on the Page
Before anything else, this is the governing principle. You should only add FAQ schema to a page if that page genuinely contains questions and answers. Marking up a page as an FAQPage when it does not have visible Q&A content is a misrepresentation, and Google's structured data guidelines are explicit about this.
This means the question is not really "every page vs. the FAQ page." The real question is: which of my pages contain actual questions and answers that users would benefit from seeing?
That could be quite a few pages, or it could be just one or two. It depends on your site structure and how you write your content.
Where FAQ Schema Genuinely Belongs
Dedicated FAQ pages
An obvious candidate. If you have a page that is built entirely around questions and answers, it absolutely should have FAQPage schema. This is the most straightforward use case, and Google tends to treat it reliably.
Service pages with embedded Q&A sections
Many well-built service pages include a section near the bottom that answers common objections or questions. "How long does delivery take?" or "Do you offer refunds?" These are real questions, with real answers, on a real page. FAQ schema is appropriate here, even though the page is primarily a service or product page.
This is actually one of the most valuable places to use it. Service pages often rank for competitive terms, and adding FAQ markup to a section at the bottom gives AI engines a ready-made Q&A block to extract when someone asks a related question.
Blog posts structured around questions
If you write a blog post that answers several distinct questions, you can mark those up using FAQPage schema. This post is a good example. The FAQ section at the bottom contains genuine questions with specific answers. That section could legitimately carry FAQ schema.
The key is that the questions and answers must appear on the page in a form users can read. You cannot hide them or only include them in the JSON-LD without showing them in the HTML.
Product pages with Q&A content
E-commerce product pages sometimes include a questions section, either pulled from customer submissions or written by the team. If the questions and answers are visible on the page, FAQ schema is valid. This can help AI systems cite your product page when someone asks "does [product] work with [something]?" type questions.
Where FAQ Schema Does Not Belong
Pages with no question-and-answer content
A homepage that just describes your services. A category page listing products. A contact page. None of these warrant FAQPage schema unless you have genuinely added a Q&A section to them. Adding the markup without the content is a guidelines violation and does not help with anything.
Pages where the "questions" are just headings
This one catches people out. Some pages use question-style headings like "Why choose us?" or "How does it work?" but do not provide a direct, self-contained answer beneath each one. That is not a Q&A structure. FAQ schema needs a genuine question paired with a complete, standalone answer that makes sense out of context.
Every page as a blanket strategy
Some SEO plugins make it tempting to apply FAQ schema globally. Resist that. Google has said explicitly that applying FAQPage schema to pages where the primary purpose is not a FAQ can result in those rich results being suppressed across your entire site. A blanket approach tends to dilute the signal and can trigger manual review.
How Many Questions Should You Mark Up?
There is no fixed maximum, but practical experience suggests keeping it between three and eight questions per page. Too few and you are barely making use of the markup. Too many and you risk the answers becoming shallow or the schema block becoming unwieldy.
Each answer should be thorough enough to be genuinely useful on its own. Think about it from an AI perspective: if Perplexity is looking for an answer to a user's question, it needs the answer to be complete without the user needing to read surrounding context. A one-sentence answer is rarely enough. Aim for at least two to four sentences per answer, covering the key point and any important caveats.
The AI Search Angle Changes the Calculus Slightly
For traditional Google SEO, the FAQ rich result is the main incentive. Google has narrowed when these appear, particularly for commercial pages, so the direct SERP benefit has reduced.
But for AI search visibility, the logic is different. AI engines like ChatGPT and Perplexity do not just look at whether a page ranks. They look at whether the content is clearly structured and machine-readable. A page with well-formed FAQ schema gives these systems a reliable, pre-structured extract to pull from when generating answers.
This means it is worth adding FAQ schema to pages that might not have been traditional FAQ candidates, as long as the Q&A content is genuinely there. A well-written service page with five specific questions marked up is more likely to get cited by an AI engine than a wall of prose that happens to contain the same information.
At FlinnSchema, we consistently find that pages with properly implemented FAQ schema outperform equivalent pages without it when it comes to AI citation rates. It is not magic. It just makes the content easier to process. Understanding why AI search needs structured data helps make sense of why this matters beyond traditional SEO.
A Practical Decision Framework
Ask yourself these questions for each page on your site:
- Does this page contain at least three questions with complete, standalone answers?
- Are those questions and answers visible in the HTML that users can read?
- Would a user reading only the question and answer get a genuinely useful response, without needing surrounding context?
- Is the primary purpose of this page something other than hiding content purely inside structured data?
If you can answer yes to all four, FAQ schema is appropriate. If you answer no to any of them, either rework the content first or leave the markup off.
For most e-commerce sites, this typically means FAQ schema belongs on: one or two dedicated FAQ or help pages, key service or category pages that include a Q&A section, and well-structured blog posts like guides or explainers. That might be ten to thirty pages on a mid-sized site. Not every page. Not just one page.
If you are unsure which of your pages are actually eligible, an AI visibility audit can identify where structured data is missing, misapplied, or working well. It is worth knowing before you start adding markup at scale.
One More Thing: Avoid Duplicate Questions Across Pages
If you are marking up FAQ schema on multiple pages, try not to repeat the exact same questions. "What is your returns policy?" appearing in FAQ schema on your homepage, your product pages, your FAQ page, and your checkout page is not adding value. It creates redundancy that AI engines and Google both notice.
Each page's FAQ schema should reflect questions that are specific and relevant to that page's topic. A product page should have product-specific questions. A service page should have service-specific questions. Your main FAQ page can cover broader or more general questions. Keep them distinct and you keep the signal clean.
For a deeper look at how different schema types work together on an e-commerce site, this guide to essential e-commerce schema types covers how FAQ fits alongside Product, Review, and Organisation markup.
Frequently Asked Questions
Can I add FAQ schema to a page that only has one or two questions?
Technically yes, the FAQPage type does not require a minimum number of questions. However, one or two questions rarely justify the markup. Google tends to only show FAQ rich results for pages with a meaningful Q&A section, and AI engines are looking for substantive content. If you only have one or two questions, consider whether the page genuinely warrants the FAQ structure or whether you should expand the content first.
Will adding FAQ schema to every page hurt my site?
It can. Google has stated that misuse of FAQ schema, including applying it to pages where it is not appropriate, can result in rich results being suppressed. There is also a risk of a manual action if the markup is seen as manipulative. Blanket application without matching on-page content is the most common mistake, and it often ends up doing more harm than good.
Does FAQ schema help with AI search engines like ChatGPT and Perplexity?
Yes, and this is increasingly one of the stronger reasons to use it. AI engines process structured data to understand what a page is about and to extract reliable answers. A well-formed FAQ section with clear, complete answers is much easier for an AI to cite accurately than unstructured prose. If you want your content to appear in AI-generated responses, FAQ schema on the right pages is a worthwhile investment.
Does the FAQ content need to be visible on the page or can it just be in the JSON-LD?
It must be visible on the page. Google's structured data guidelines require that the content marked up in schema is actually present and readable in the page's HTML. Hiding answers inside JSON-LD without showing them to users is a clear violation and can result in penalties. The markup should describe the content, not replace it.