Google Killed FAQ Rich Results (May 2026): The Hot Take That Replaces Them Is Wrong Too

What just happened: Google killed FAQ rich results

On May 7, 2026, Google quietly stopped showing FAQ rich results in Search. The expandable Q&A dropdowns under listings — the ones that let users read three or four questions before clicking through — are gone for every site.

The official update appeared on the FAQPage structured data documentation the next day. The opening line now reads, in plain text: “FAQ rich results are no longer appearing in Google Search.” That single sentence formally ended a feature Google launched in May 2019, peaked in usage in 2020–2022, and largely restricted in August 2023.

Google also confirmed three follow-up steps. The FAQ search appearance, the rich result report inside Search Console, and the Rich Results Test support all turn off in June 2026. Search Console API support for FAQ data ends in August 2026.

Google’s guidance to site owners is short: “While you can drop this structured data from your site, there’s no need to proactively remove it. Structured data that’s not being used does not cause problems for Search.” Translation: don’t bother deleting your markup.

A six-year wind-down, not a sudden update

If you read the headlines this week, this looked sudden. It wasn’t. Google has been telegraphing this exact outcome since 2020. Anyone who says they were caught off guard hasn’t been watching the signals.

Here is the full timeline, stage by stage.

2020 — The first visible pullback

As early as August 2020, the SEO community noticed FAQ rich results appearing on fewer queries. Google’s John Mueller acknowledged it openly, saying the team was “trying to find the right balance between showing these everywhere and showing these for pages where it kind of makes more sense.” That was the first public admission that the rich result was being throttled.

June 2021 — Capped at two FAQs per page

In June 2021, Google launched what the community called the “FAQ review update”. The cap dropped from up to 10 FAQs per page to a hard maximum of two — an 80% reduction in SERP real estate available to FAQ markup. Google Search Liaison Danny Sullivan confirmed the change on Twitter.

Notably, Sullivan publicly pushed back on the idea that the cap was an anti-abuse move. He wrote: “I’ve seen several people suggest this is something we did because of SEO gaming. It’s not. SEOs couldn’t game how many we displayed… we just review things all the time and determined showing up to two was most useful.” Whatever the framing, the practical effect was the same: fewer slots, less incentive to add long FAQ blocks.

April 2023 — A quiet rollback

Months before the big public announcement, Google reduced FAQ rich result visibility through a Search Console notification only. No blog post. Many site owners just saw their FAQ impressions fall and didn’t know why for weeks.

August 8, 2023 — Restricted to health and government sites

This is the one most readers will remember. Google’s official Search Central post announced that FAQ rich results would only be shown for “well-known, authoritative government and health websites”. The change rolled out globally within a week.

For 99% of sites on the web, this was the real death of the feature. The May 2026 announcement just made it official for the remaining 1%.

Independent SEO consultant Brodie Clark documented the rollout in real time, tracking GSC properties for large sites that used FAQ schema heavily.

Brodie Clark

@brodieseo

·Follow

The day has come. Google has now removed FAQ Schema rich results for the vast majority of sites on desktop. I checked a few GSC properties for large sites that use it heavily, and its gone. This is over a month since the change was meant to happen, but we knew it was coming.

11:42 AM · Sep 13, 2023

163

Reply

Copy link

Read 9 replies

September 13, 2023 — HowTo killed across desktop too

Five weeks later, Google extended the HowTo deprecation to desktop, fully retiring that sister rich result type. If you were watching, this was the writing on the wall: structured-data rich results that depend on a single SERP feature were being shut down one by one.

March 2026 — The final 50% drop

For the small set of gov/health sites still eligible, FAQ rich result impressions fell by roughly half compared to the post-2023 baseline. Trackers like Search Engine Land flagged it. The May full removal was the obvious next step.

May 7, 2026 — The chapter closes

Full deprecation. Every site, every query.

The full sequence: gradual reduction (2020) → hard limit to 2 (2021) → quiet rollback (April 2023) → restriction to gov/health only (August 2023) → complete deprecation (May 2026). Anyone who was surprised at any point after 2021 was not reading the signals.

Why did Google do it now?

Google didn’t give a public reason. Two patterns explain it well.

1. The feature was widely abused. Marketers used FAQ schema to grab extra SERP real estate, often with low-quality or irrelevant questions. Google didn’t always frame the rollbacks as anti-abuse measures — Sullivan denied the 2021 cap had anything to do with gaming — but by August 2023 the trajectory was clear: restrict the rich result to “well-known, authoritative” sites only. Six patterns showed up in almost every audit.

Search Engine Land’s retrospective put it bluntly: the feature was “stuffed with keyword-heavy questions, irrelevant content, or duplicated information that didn’t genuinely help searchers”. Once that pattern hit a critical mass of the SERP, Google had to step in.

2. Google is reducing structured-data SERP surface area in general. HowTo died in 2023. The Sitelinks Search Box was retired globally on November 21, 2024 because usage had quietly dropped over a decade. FAQ is now the third major rich result to be killed in three years.

Lily Ray, one of the most watched voices in the SEO community, flagged the timing publicly the day after the change.

Lily Ray 😏

@lilyraynyc

·Follow

Interesting timing: Google is dropping rich results support for *all* FAQ Schema as of yesterday, and is removing FAQ structured data reporting from GSC. (Link in comments) I wonder why they decided to do this *right now* in May 2026? Google had already deprecated FAQ rich Show more

4:31 PM · May 8, 2026

221

Reply

Copy link

Read 23 replies

The implied answer in her thread: this isn’t just about FAQ schema. It’s about Google reducing the surface area of SERP features at the same time AI Overviews are eating into traditional listings. Less SERP real estate means less to optimise for the old way — and more pressure to compete on content quality.

The popular narrative — and why it’s half-wrong

Within 48 hours of the announcement, the SEO blog ecosystem produced a very consistent take: “FAQ rich results are dead in Google Search, but FAQ schema now matters more than ever for AI search.”

Specific stats are flying around. You’ll see claims like “3.2× more likely to appear in AI Overviews” or “36% more likely to be cited” or “89% correlation with Perplexity citations”.

The story is comforting. It says: don’t worry, your FAQ schema work wasn’t wasted — the value just shifted from Google to AI engines.

The story is also half-wrong. Most of those stats trace back to vendor blog posts citing each other. The actual controlled studies tell a more complicated story.

What controlled studies actually say about FAQ schema and AI

Here are the studies that have run actual experiments, not surveys or correlations from small samples.

Foglift, December 2024

The Foglift research team analysed schema coverage across thousands of pages and matched it against LLM citation frequency. The finding: no statistically meaningful correlation between schema markup coverage and LLM citation frequency. More schema does not equal more citations. (See Sources below for the original study.)

SE Ranking dataset

SE Ranking analysed 129,000 unique domains across 216,524 pages in 20 niches and matched FAQ-schema use against ChatGPT citation counts. Pages with FAQ schema averaged 3.6 citations. Pages without it averaged 4.2. A slight negative correlation — the opposite of what the popular narrative claims.

The same study found referring-domain count was the single strongest predictor of being cited, with a threshold effect at around 32,000 referring domains where citations nearly doubled. One plausible read: FAQ schema is disproportionately used on lower-authority pages, which is why it correlates with fewer citations — not because the schema hurts.

The Williams-Cook experiment, February 2026

This is the highest-rigor test publicly available, documented by ZipTie’s analysis of FAQ schema and AI citations. Researchers planted fake addresses in pages, but only inside invalid JSON-LD blocks. Both ChatGPT and Perplexity successfully extracted the fake addresses anyway.

The implication: LLMs tokenise JSON-LD as raw text. They don’t parse it as a structured data layer. The schema validation that matters to Google is invisible to the LLM — it just reads the words.

Nature Communications, February 2024

The other side of the picture. A study published in Nature Communications showed that LLMs extract information more accurately from structured fields than from unstructured prose. So when an LLM does encounter your content, schema can help it parse facts cleanly.

Data World benchmark

A Data World benchmark found GPT-4’s accuracy on Q&A tasks jumped from 16% to 54% when content was backed by structured data. Again: structure helps quality, not selection.

Who has actually confirmed they use schema?

The honest answer: only Google and Microsoft. Microsoft has publicly confirmed Bing and Copilot use schema.org markup to understand pages. Google has long said the same for Search.

OpenAI, Anthropic, and Perplexity have not publicly disclosed whether they use schema during indexing or citation selection. For three of the four biggest AI search players, we’re in the dark. Most blog posts that confidently state “ChatGPT loves FAQ schema” are guessing.

The real mechanism: indirect, through Google’s pipeline

Stitching the evidence together gives a clearer picture. FAQ schema can help your AI visibility, but the path is indirect.

1. Schema feeds Google’s Knowledge Graph — cleaner entity data, better understanding of your page.
2. Knowledge Graph and Search rankings feed AI Overviews — pages that rank well in normal Search are more likely to be cited in AI Overviews.
3. AI Overviews citations feed brand visibility — users see your name even when they don’t click.

That’s a real value chain. It just isn’t the “add FAQ schema and ChatGPT will cite you” story most blog posts are selling.

What good FAQ markup looks like in practice

Even with rich results gone, the same code rules still matter for AI extraction and Knowledge Graph signals. The difference between markup that helps and markup that hurts comes down to one thing: does the visible page match the JSON-LD?

Here is a minimal valid FAQPage block, paired with the visible HTML it should reflect.

<!-- Visible on the page (raw HTML, no JS required) -->
<section>
  <h2>Frequently asked questions</h2>
  <h3>How long does shipping take?</h3>
  <p>Standard shipping arrives in 3–5 business days
     in the continental US.</p>
  <h3>Can I return an item?</h3>
  <p>Yes, returns are accepted within 30 days of
     delivery for a full refund.</p>
</section>

<!-- Matching JSON-LD in the head or body -->
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "How long does shipping take?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "Standard shipping arrives in 3–5 business days in the continental US."
    }
  }, {
    "@type": "Question",
    "name": "Can I return an item?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "Yes, returns are accepted within 30 days of delivery for a full refund."
    }
  }]
}
</script>

<!-- Page body has zero FAQs visible to a user -->
<main>
  <h1>Buy Wireless Headphones – $89</h1>
  <button>Add to cart</button>
</main>

<!-- But the JSON-LD claims 5 questions exist -->
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "What is the best cheap headphones to buy
             online affordable bluetooth wireless 2026?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "Call us at 1-800-BUY-NOW for 50% off our
               premium model! Limited time only!"
    }
  }, /* ...4 more equally fake questions... */]
}
</script>

The bad version breaks three rules at once: (a) content marked up isn’t visible to users (a guidelines violation), (b) the question text is keyword-stuffed and unnatural, and (c) the answer is promotional. Any one of these alone is enough to disqualify the markup; together they describe most pages that lost eligibility in 2023.

The bigger pattern: SERP-feature SEO is brittle by design

Step back from FAQ schema for a moment. In the last three years, Google has gutted or killed three structured-data features that SEOs built strategies around.

The pattern is the same in every case:

1. Google introduces a SERP feature.
2. SEOs and tools optimise for it.
3. Adoption scales, abuse follows.
4. The feature gets cluttered or low-value.
5. Google restricts or kills it.
6. Sites that built strategies around triggering that feature lose their advantage overnight.

The lesson is uncomfortable but useful: any SEO tactic whose value depends on a single SERP feature is brittle by design. That doesn’t mean you should stop using rich results — it means rich results should be a bonus, not the foundation of your strategy.

What to actually do now

Here is the practical, no-hype version. Most of it is dull. That’s the point.

What about other schema types? Are they next?

FAQ and HowTo are gone. Sitelinks Search Box is gone. Reasonable question: which schema types are safe and which look risky?

The honest answer: the schema types tied to commerce, news, and identity are the safest. Anything that gives Google direct revenue or directly fights misinformation is well-defended.

First, one piece of housekeeping that trips up almost everyone: FAQPage and Q&APage are not the same schema, and only one of them is dead. FAQPage is what just got deprecated — it’s for pages where you, the site owner, provide a single authoritative answer to each question. Q&APage is for community-style pages where users post a question and other users post competing answers (Stack Overflow, Quora, support forums). Q&APage is still fully supported. If your support forum or community thread uses Q&APage, leave it alone.

How to check your site in 10 minutes

For 99% of sites, FAQ rich results have been invisible since August 2023, so there’s nothing to mourn there. The useful work is making sure the schema you do rely on is healthy. Here’s a 10-minute audit, in priority order.

1. Confirm your visible Q&A renders without JavaScript. Open any page that uses FAQPage markup, disable JavaScript in your browser, and reload. If the questions and answers vanish, AI engines probably can’t tokenise them either — this is the single highest-leverage fix in this article.
2. Validate every active schema type. Run your homepage and 3–5 high-traffic templates through our free Schema & JSON-LD validator — it checks JSON syntax, required properties, date formats, and nested entities for Article, Product, Organization, BreadcrumbList, FAQPage, and 8 other types. Look for warnings on the schemas that still render and feed Knowledge Graph.
3. Add or strengthen author schema on YMYL pages. If your finance, health, or legal content doesn’t link Article.author to a real Person with credentials, fix that today. The March 2026 core update measurably rewards verified authorship.
4. Pull historical FAQ data from GSC only if you’re a gov/health site. Open Google Search Console → Performance → Search results, click + New, choose Search appearance, and look for FAQ in the dropdown. If it appears, export the last 16 months before June 2026. If it doesn’t appear, your site hasn’t recorded FAQ rich-result data since 2023 — skip this step.
5. Audit the rest of your structured data at scale. Manual checks cover the homepage. A full crawl surfaces every page with deprecated schema, validation errors, or missing fields — the issues that compound across thousands of URLs. CrawlRaven’s 200-point audit validates structured data on every crawled page automatically.

For the broader context on how this fits with other recent Google updates, see our analysis of the March 2026 core update and the March 2026 spam update — both reinforce the same theme: SEO that depends on visible SERP features is increasingly fragile.

Bottom line

Google killed FAQ rich results on May 7, 2026. The mainstream story — “dead in Google Search but alive for AI” — is half right and half wishful thinking.

The full picture is more honest. The deprecation was telegraphed for six years, so nobody should be surprised. The popular “FAQ schema is now an AI citation magnet” claim doesn’t hold up to the studies that have actually been run. And the structural lesson — that SEO tactics tied to a single SERP feature are brittle — is now backed by three deprecation events in a row.

Practical advice: leave the schema in place, write Q&A content that’s actually useful, and stop using Rich Results Test as a checklist gate. Boring fundamentals beat clever tricks — that’s the throughline of every Google update for the last decade, and it’s the throughline of this one too.

Sources and further reading

• Google — Mark Up FAQs with Structured Data (updated May 8, 2026)
• Google Search Central — Changes to HowTo and FAQ rich results (Aug 2023)
• Search Engine Land — Google to no longer support FAQ rich results
• Search Engine Land — The rise and fall of FAQ schema
• Search Engine Journal — Google Downgrades Visibility of HowTo and FAQ Rich Results
• Vizup — Google Kills FAQ Rich Results: What It Means for AEO in 2026
• faqjsonld.com — Google FAQ Schema Deprecation in 2026
• ZipTie — FAQ Schema for AI Answers: Does It Actually Get You Cited?
• Lily Ray on X — on the May 2026 FAQ deprecation
• Brodie Clark on X — FAQ rich result rollout (2023)
• Seer Interactive — FAQ Reduced Visibility & No More HowTo
• Search Engine Roundtable — Google Sitelinks Search Box Removal (Nov 2024)
• CrawlRaven — Google March 2026 Core Update: Impact & Recovery
• CrawlRaven — Google March 2026 Spam Update: Fastest Rollout Ever
• CrawlRaven — How to Perform a Technical SEO Audit in 2026