Your FAQ enhancement report vanished from Search Console, and now you are wondering whether the markup was ever worth the effort. Google retired FAQ rich results on 7 May 2026, and the SEO industry split into two loud camps inside a day. This guide covers how to add FAQ schema for AI Overviews, what the evidence says it really does, and whether your existing markup should stay. You get a working template, install steps for four platforms, and a validation routine that survives the June 2026 tool changes.
To add FAQ schema for AI Overviews, write real questions with short visible answers, wrap them in FAQPage JSON-LD, paste the script into your page head, then validate with the Schema Markup Validator. Google retired FAQ rich results in May 2026, so the visible question-and-answer content now does the heavy lifting.
One thing needs settling before any code. FAQ schema and FAQ rich results are two different things, and almost all the May panic came from treating them as one. Affordable SEO Services put this walkthrough together after that documentation change, and the order below is deliberately the order you should work in.
What Is FAQ Schema and Does It Still Work in 2026?
FAQ schema is a block of code that labels which parts of your page are questions and which parts are the answers. It uses the FAQPage type from Schema.org, and each pair sits inside a Question object with an acceptedAnswer attached to it.
Nothing about it changes what a visitor sees. The markup is a translation layer for machines, not a design feature, and that distinction explains most of what follows.
You could write the same semantic markup as microdata, using attributes sprinkled through your HTML, but Google recommends JSON-LD and every tool in this guide assumes it. Microdata is harder to maintain and easier to break during a template change.
Does it still work?
That depends on what you mean by work. As a Schema.org type, it is completely valid, and Google still documents it. As a way to win rich snippets and extra space in the results, it is finished.
What survives is the underlying job. Question and answer pairs are the most extractable content shape on the web, which is why featured snippets, voice search, and AI Overviews all reward the same formatting.
What Changed When Google Dropped FAQ Rich Results?
Google added a deprecation notice to the top of its FAQ structured data documentation on 7 May 2026, and FAQ rich results stopped appearing in Search. There was no blog post and no webmaster announcement, just a small label on a developer page. Within a day, one half of the industry declared FAQ schema dead, and the other half declared it more important than ever.
Both readings were wrong, because each one confuses the feature with the format.
Which three dates matter?
- 7 May 2026. FAQ rich results no longer appear in Google Search for every site.
- June 2026. The FAQ search appearance, the rich result report, and FAQ support inside the Rich Results Test were all dropped.
- August 2026. Support for the FAQ rich result in the Search Console API is removed.
If any dashboard or Looker Studio connector pulls FAQ appearance data through that API, the pull is about to break. Archive the historical numbers this week.
Why did Google pull the feature?
Google gave no public reason, but its own three-year pattern tells the story. In August 2023, the company restricted FAQ rich results to well-known, authoritative government and health sites, which removed eligibility for almost every commercial page. HowTo rich results went from desktop the following month. Seven more structured data types were retired in June 2025, including Course Info, Claim Review, and Estimated Salary, on the reasoning that they were not widely used and no longer added much for users.
The thread running through all of it is abuse. Once a rich result gets bolted onto every page by a plugin, it stops describing the page and starts decorating the listing.
Should you delete your existing FAQ markup?
No. Google has said unused structured data does not cause problems for Search, and the May documentation update specifically notes the markup can be left in place. There is no penalty for keeping valid FAQPage code where it is.
What it does cost you is maintenance. If your markup no longer matches the visible page, that is a reason to fix it or pull it, and it always was.
Does FAQ Schema Actually Get You Cited in AI Overviews?
Probably not on its own. The strongest causal test available found no meaningful citation uplift from adding schema at all, which is the opposite of what most guides on this topic will tell you.
Start with the number everyone quotes. Ahrefs analysed 6 million URLs and found that pages cited by AI were roughly three times more likely to carry JSON-LD. Frase, citing an SGA Index analysis, puts FAQ schema pages at 3.2 times the citation rate of pages with no structured data.
Correlation was never really the question, though. Sites that bother with schema also publish better content, earn more links, and maintain their templates properly, so the markup may simply be a marker for everything else they do well.
Ahrefs went and ran the harder test. Louise Linehan and Xibeijia Guan tracked 1,885 pages that added JSON-LD for the first time between August 2025 and March 2026, matched them against 4,000 control pages on other domains with similar prior citation levels, and ran a difference in differences analysis across Google AI Overviews, AI Mode, and ChatGPT. Google AI Mode moved 2.4 percent. ChatGPT moved 2.2 percent. Both sit inside the noise.
AI Overviews came in at minus 4.6 percent, small but statistically real. A nick, not a crater, and in the wrong direction. Their own summary of it was blunt. Adding schema produced no major uplift in citations on any platform.
A separate experiment from searchVIU asked a narrower question. It tested whether five major AI systems read schema when fetching a page live, and ChatGPT, Claude, Perplexity, Gemini and Google AI Mode all pulled only the visible HTML. JSON-LD, hidden Microdata and hidden RDFa were ignored at the point of retrieval.
Mark Williams-Cook demonstrated the same thing from another angle. He published a test page with deliberately broken schema; the models returned the data anyway, and a good chunk of the industry read that as proof they were parsing his markup. They were reading the HTML around it.
So why bother adding it at all?
Because the Ahrefs study has limits its own authors are open about. It measured what happens when you add schema to pages that AI systems already cite. It cannot tell you what happens on a page nothing is looking at yet, where markup may still help with crawlability, parsing, and entity resolution.
There is also the crawler layer to think about. Bingbot, PerplexityBot, and the retrieval crawlers feeding LLM systems all fetch your JSON-LD along with everything else. How any of them weight it in answer generation is undocumented, so nobody can honestly tell you it is worthless either.
Here is the practical read. Write the question and answer content because the content is what gets extracted. Add the markup because it is cheap, it is valid, and it keeps your page legible to the many Google systems that have nothing to do with AI. Just stop selling it internally, or to a client, as a citation lever.
How Do You Add FAQ Schema for AI Overviews Step by Step?
Work in this order. Jumping straight to step three is how most broken implementations happen.
Step 1. Choose questions people actually search
Pull questions from Search Console query data, People Also Ask boxes, and autocomplete suggestions. Every question should be one a real person types, phrased the way they type it.
Never invent questions to fill a template. Generic filler along the lines of “Why choose us?” is exactly the pattern that got the rich result killed.
Step 2. Write answers a machine can lift cleanly
Answer in the first sentence, then expand for one or two more. Industry convention puts the extractable answer at 40 to 60 words, although no platform has ever published a rule, so treat that as a guide rather than a spec.
Keep one answer per question. Strip promotional language out, because a sales pitch inside an Answer object breaches Google’s guidelines and reads badly to humans as well.
Step 3. Build the JSON-LD, with a working example
Copy the template below and swap the placeholder text for your own. Keep the structure exactly as it is.
<script type=”application/ld+json”>
{
“@context”: “https://schema.org”,
“@type”: “FAQPage”,
“mainEntity”: [
{
“@type”: “Question”,
“name”: “Your first question, exactly as it appears on the page”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “Your answer, matching the visible page text word for word.”
}
},
{
“@type”: “Question”,
“name”: “Your second question, exactly as it appears on the page”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “Your second answer, again matching the visible page text.”
}
}
]
}
</script>
Step 4. Install it on your platform
Pick the route that matches your stack.
- WordPress. Generate the block through Rank Math or Yoast SEO if your version outputs FAQPage, or paste the script into the head using a code snippets plugin.
- Shopify. Paste the script into the theme. liquid before the closing head tag, or into the specific template file when the FAQs are page-specific.
- Webflow. Open Page Settings for that page and drop the script into the custom code head section.
- Google Tag Manager. Deploy a Custom HTML tag on the matching page URL. Use this only when you cannot edit the template, because a tag manager adds a rendering dependency between your markup and a crawler.
Wherever the code ends up, it has to match the visible text on the page word for word.
Step 5. Validate before you publish
Run this list every time, in this order. It takes four minutes and catches every failure mode in the next section.
- Paste the URL into the Schema Markup Validator at validator.schema.org, which checks your syntax against Schema.org rather than against Google features.
- Skip the Rich Results Test for this type. Google dropped FAQ support from it in June 2026, so a clean pass there now tells you nothing.
- View source on the live page and confirm the script actually rendered. Schema injected by JavaScript or a tag manager sometimes never reaches the crawler at all.
- Read the visible answers against the markup, word for word. Any drift between the two is a guideline breach.
- Paste the published URL into an LLM and ask which questions the page answers. If it returns anything other than your FAQs, your visible content is the problem rather than your code.
Record your Search Console impression baseline and note the date before the change goes live.
Step 6. Measure it in the new Search Console reports
Google launched dedicated generative AI performance reports in Search Console on 3 June 2026, announced by Hillel Maoz and Moshe Samet. They separate impressions from AI Overviews and AI Mode out of the general performance report for the first time.
Know what you are getting, though. The reports show impressions, pages, countries, devices, and dates, with no clicks, no click-through rate, no query data, and no API access. History starts at roughly 18 May 2026, and the rollout is still incremental, so not every property has the view yet.
Record a 30-day impression baseline before you touch a template. Affordable SEO Services treats this as non-negotiable on client sites, because without a baseline you cannot separate a schema effect from a core update landing the same week.
How Do You Layer FAQPage With Article and Organization Schema?
FAQPage rarely lives alone. A blog post carrying an FAQ block usually needs Article schema, an Organization entity for the publisher, a Person entity for the author, and a BreadcrumbList for navigation. That is five types on one page, and how you arrange them matters more than most guides admit.
Two arrangements work, and Google processes both. Give each entity its own script block, or wrap everything in a single block using an @graph array. Separate blocks suit entities with no relationship to each other. The @graph array suits pages where the entities connect, which describes nearly every article page you will ever mark up.
The mechanism that makes layering worth the effort is @id. Assign each entity a unique identifier, then reference that identifier elsewhere instead of repeating the whole object. Yoast’s developer documentation puts the case plainly, since a product page does not need to repeat all the markup for the organization selling the product when it can point at that organization by ID.
One source of truth, no contradictions between blocks, and a much smaller maintenance surface when the company name or logo changes.
A layered example
<script type=”application/ld+json”>
{
“@context”: “https://schema.org”,
“@graph”: [
{
“@type”: “Organization”,
“@id”: “https://example.com/#organization”,
“name”: “Your Company”,
“url”: “https://example.com/”
},
{
“@type”: “Article”,
“@id”: “https://example.com/guide/#article”,
“headline”: “Your article title”,
“inLanguage”: “en”,
“publisher”: { “@id”: “https://example.com/#organization” },
“author”: {
“@type”: “Person”,
“name”: “Author Name”,
“jobTitle”: “Technical SEO Lead”
}
},
{
“@type”: “FAQPage”,
“@id”: “https://example.com/guide/#faq”,
“mainEntity”: [
{
“@type”: “Question”,
“name”: “Your question, exactly as it appears on the page”,
“acceptedAnswer”: {
“@type”: “Answer”,
“text”: “Your answer, matching the visible page text.”
}
}
]
}
]
}
</script>
Where does E-E-A-T fit into this?
Not as something you can mark up directly, because Schema.org has no experience property and no expertise score. What layering does is make the author and the publisher explicit instead of leaving Google to infer them from a byline. That is entity clarity, and it is a separate job from the FAQ pairs themselves.
A warning before you build any of it. Rank Math and Yoast SEO already emit their own @graph containing Organization, WebSite, WebPage, BreadcrumbList, and Article nodes. Bolt a handwritten block on top, and you get schema collision, with two Organization entities describing the same company in slightly different words. Check what your plugin outputs before you write a line.
The honest note belongs here too. No controlled study shows that layering causes AI citations any more than a plain FAQPage block does, and the vendor case studies claiming triple-digit citation gains from nesting are not controlled tests. Layer your schema for machine clarity and internal consistency, which are real, and not for a number somebody put in a deck.
Does Every AI Platform Read FAQ Content the Same Way?
No, but the differences matter far less than most guides suggest. Google is the only one of the three with a documented structured data pipeline. ChatGPT and Perplexity have published nothing about how they weight schema, or whether they weight it at all.
Recall what the searchVIU test found. During live retrieval, every system it examined took the visible HTML and left the JSON-LD alone. That points at one formatting rule rather than three separate playbooks.
Google is still the exception worth planning around, because Google is not one system. The index, the ranking systems, the Knowledge Graph, and AI Overviews are separate pieces that between them produce something that looks like a single answer. Structured data feeds the parts of that stack that existed long before AI Overviews did.
Formatting that travels well across all of them looks identical to formatting for a featured snippet. Direct answer first, roughly 40 to 60 words, one idea per pair, plain vocabulary and no throat clearing. Voice search rewards the same shape, which is no coincidence, because every one of these surfaces is trying to lift one clean passage out of your page.
How Do You Handle FAQ Schema on Multilingual Sites?
Give every locale its own schema block, written in that locale’s language. International teams get this wrong constantly, usually by translating the visible page and leaving English markup sitting underneath it.
Three rules cover nearly every case.
- Set inLanguage on each block to match the page, using the same language code your hreflang tags use, so the two signals agree with each other.
- Write the question and answer text in the page language. English markup on a Spanish page is a mismatch between code and visible content, which is the first mistake from the checklist above wearing a different coat.
- Point every URL inside the schema at the locale version rather than the English default, including any @id values you use for layering.
Hreflang and schema do different jobs, and neither one replaces the other. Hreflang tells Google which URL to serve to which audience. The markup tells it what the page means. They only cause trouble when they disagree.
How Many FAQs Should One Page Have?
Four to eight on most pages. Past that, the returns fall away fast, and three specific problems start showing up.
Cannibalisation is the first. Fifteen questions on one page usually means several of them deserve a page of their own and are now competing with the page they sit on.
Thin duplication is the second. When the same FAQ block gets stamped across an entire template, none of those pages is answering anything specific to itself.
Dilution is the third. Extraction systems pick the pair that best matches the query, and a long list simply gives them more ways to pick the wrong one. Some practitioners cap it at six for that reason, though no platform has published a threshold anyone can point at.
Which Five Mistakes Silently Break FAQ Schema?
Each of these invalidates the markup or wastes it, and none of them shows up as an error on the page.
- Markup that does not match the visible page. If the answer only exists in the code, it breaches Google’s guidelines and no retrieval system will see it anyway.
- Promotional answers. Offers, pricing pushes, and calls to action inside an Answer object are a guideline breach.
- Duplicate questions across pages. The same pair repeated site-wide gives every page an identical signal, which is the same as no signal.
- More than one answer per question. FAQPage expects a single official answer, so multiple answers means you want QAPage or a restructured page.
- A missing comma or bracket. One syntax error invalidates the entire block silently, which is why step five is not optional.
FAQ Schema or Q and A Schema, Which Should You Use?
Use FAQPage when you write the questions, and you provide the one official answer. Use QAPage when users submit the question and other users post competing answers, the way a forum thread or a support community works.
Google still documents both types and still treats them as separate things, so the May deprecation did not merge them. Getting this wrong during a template audit is an easy way to strip valid markup you meant to keep.
So start with the questions, not the code. Open Search Console, sort your queries by impressions, and pull out every one phrased as a question. Answer the five best on the page where they belong, then add the markup underneath. If you would rather have that audited across a whole site than one page at a time, Affordable SEO Services can take the technical side off your hands.
Frequently Asked Questions
Is FAQ schema dead in 2026?
No. The FAQ rich result is gone from Google Search, but FAQPage remains a valid Schema.org type that Google still documents.
Should I remove FAQPage markup from my site?
No. Google has said unused structured data does not cause problems for Search, so valid markup can stay where it is.
Does FAQ schema help you get cited in AI Overviews?
The evidence says not on its own. Ahrefs tracked 1,885 pages that added schema and found no meaningful citation uplift on any platform.
How long should each FAQ answer be?
Roughly 40 to 60 words for the direct answer, then expand if the question needs it. This is industry convention, not a published platform rule.
Can FAQs sit inside an accordion and still be marked up?
Yes, provided the answer text is in the page HTML and a visitor can open it. Content a user can never reach is a guideline problem.
How do I test FAQ schema now the Rich Results Test dropped support?
Use the Schema Markup Validator at validator.schema.org, which checks your syntax against Schema.org rather than against Google features.
What is the difference between FAQPage and QAPage?
FAQPage is for your own single official answer. QAPage is for pages where users submit multiple competing answers to one question.
How many FAQs should be on one page?
Four to eight is a sensible working range. More than that risks cannibalising pages that deserve to stand on their own.
Can you combine FAQ schema with Article and Organization schema?
Yes. Wrap them in a single @graph array and link the entities with @id references so nothing is duplicated or contradicted.
Should FAQ schema be translated on multilingual pages?
Yes. Each locale needs its own block written in that language, with inLanguage matching your hreflang code and locale-specific URLs.
Is FAQ schema still worth adding for AEO and GEO in 2026?
As supporting structure, yes. As a standalone answer engine optimisation tactic, no, because the visible content is what gets extracted.
What Should You Do Next?
Knowing how to add FAQ schema for AI Overviews matters far less than knowing why you are adding it. The markup was never doing the work. The content was, and Google removing the rich result only made that obvious to everyone at once.
