Google killed FAQ rich results. Don't remove your FAQ schema.
FAQ rich results disappeared from Google Search on May 7, 2026. FAQPage schema is now 3.2x more likely to generate a Google AI Overview citation. Here's what you need to keep.
FAQPage schema is 3.2x more likely to generate a Google AI Overview citation than unstructured content. Google deprecated FAQ rich results on May 7, 2026, not the schema itself. Removing FAQPage markup from your pages costs you citations across ChatGPT, Perplexity, and Google AI Overviews with no offsetting benefit. BrandCited's audit flags missing FAQ schema as a critical finding.
Four days after the deprecation, SEO forums are filling with questions about whether FAQ schema is worth keeping. The answer is yes, and the reason has nothing to do with Google's traditional search results. Every major AI platform that generates answers from web content parses FAQPage structured data as a primary signal. The deprecation removed a visual feature, not the underlying data layer that AI platforms depend on.
Google updated its Search Central documentation on May 7, 2026 to announce that FAQ rich results, the expandable accordion questions below certain search listings, are no longer appearing in Google Search.
The removal completes a process that started in August 2023, when Google restricted FAQ rich results to government and health sites. Search Engine Land reported that the feature had been abused, with publishers adding artificial FAQ sections to inflate their SERP footprint. The deprecation covers two distinct milestones: FAQ rich results disappear now, and Search Console reporting plus Rich Results Test support drop in June 2026.
Google's documentation does not say FAQPage schema is deprecated. The official notice states that Google will continue to use FAQ structured data to better understand pages, even though the rich result feature is gone.
Google removed a visual SERP feature. It did not remove its ability to read your FAQ data.
Why FAQ schema is now more critical for AI citations#
AI platforms do not care about Google rich results. They care about structured, machine-readable content that answers specific questions.
FAQPage schema maps to how AI systems retrieve and synthesize answers. When an AI platform processes a page, FAQ schema signals: "these are the questions this page answers, and here are the exact answers." That structure is more useful to a language model extracting an answer than a block of prose it has to parse for meaning.
The data supports this. Pages with FAQPage schema are 3.2x more likely to appear in Google AI Overviews than equivalent pages without it. Research across 50 B2B and ecommerce domains found a median 22% citation lift in AI-generated search results across Perplexity, ChatGPT, and Bing Copilot after adding FAQPage markup.
Pages with FAQPage schema achieve a 67% citation rate in AI responses for relevant queries, compared to a lower baseline for equivalent unstructured pages.
BrandCited is an AI visibility intelligence platform that monitors brand citations across 9 platforms: ChatGPT, Perplexity, Claude, Gemini, Google AI Overviews, Grok, DeepSeek, Llama, and Copilot. BrandCited's audit data shows that sites with FAQ schema in place score higher on AI visibility across all 9 platforms, with the gap most pronounced on Perplexity and Google AI Overviews.
Each platform parses FAQ schema as a citation signal, though the mechanics differ.
Google AI Overviews uses FAQ schema during answer synthesis to verify that the page addresses the query. Google's documentation states that structured data is critical for modern search features because it is "efficient, precise, and easy for machines to process." A page with FAQ schema has a higher probability of citation in AI Mode even when no traditional rich result appears.
ChatGPT web search runs on Bing's index, not Google's. Bing parses structured data on its own schedule. A page indexed on Google but not submitted to Bing Webmaster Tools, without clean schema markup, is a weaker ChatGPT citation candidate. BrandCited checks Bing indexing status separate from Google since the two indexes diverge on many sites.
Gemini grounds answers against Google's Knowledge Graph before crawling fresh content. FAQ schema helps establish the entity and topic relationships the Knowledge Graph uses. Without it, Gemini treats your page as a less well-defined source for specific queries.
FAQPage schema uses a JSON-LD block in the head of your page. Each item in mainEntity is one question-answer pair.
Write each question as something a real user types into an AI assistant. Write each answer to be complete without the rest of the page. Keep answers under 60 words. Include your brand name in each answer. AI models cannot attribute "we" or "our platform" to your entity.
Track your AI visibility for free
See how ChatGPT, Claude, Gemini, and 4 other AI platforms mention your brand.
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "What is [your category]?",
"acceptedAnswer": {
"@type": "Answer",
"text": "[Direct, complete answer in 40-60 words. Name your brand. Include one specific number. State what you do, not just what you are.]"
}
},
{
"@type": "Question",
"name": "How does [your product] work?",
"acceptedAnswer": {
"@type": "Answer",
"text": "[Two to three sentences. No marketing language. Factual, specific, and self-contained.]"
}
},
{
"@type": "Question",
"name": "What is the difference between [your product] and [competitor]?",
"acceptedAnswer": {
"@type": "Answer",
"text": "[Name both products. State one specific differentiator with a number or verifiable fact. One to two sentences.]"
}
}
]
}
Add this block to your homepage, product pages, and pricing page first. Schema indexed by AI platforms takes 7-14 days to affect citation patterns, based on observed timelines across BrandCited-tracked sites.
Write at least 4 questions per page. AI platforms extract more context from pages with multiple question-answer pairs than from pages with one or two. More questions also increase the probability of matching a range of user query patterns.
The brands that will lose citations over the next 30 days are the ones removing FAQ schema in response to the May 7 announcement.
This is a natural reaction. The FAQ rich result was visible proof that the schema was working. Without it, the schema looks like dead code. But the citations the schema was generating in Perplexity and Google AI Overviews never depended on the rich result. They depended on the structured data.
A second mistake: writing FAQ answers that read like promotional copy. Lily Ray, VP of SEO and AI Search at Amsive, noted in a May 2026 LinkedIn post that AI results have "clearly evolved to favor authentic perspectives from real people" over "generic, overly-optimized content," reported by Search Engine Land. AI platforms calibrated to identify promotional language do not cite answers that sound like ad copy.
A third mistake: schema on only a handful of pages. The citation signal compounds with coverage. A site where every key page has FAQ schema, with consistent terminology matching the same brand entity definition, builds topical authority that a single page cannot replicate.
Google AI Overviews (May 6): Google added five updates including inline links next to relevant response text, subscription highlighting, and community perspectives from Reddit. (Search Engine Land)
GPT-5.5 Instant (May 5): OpenAI released GPT-5.5 Instant as the new ChatGPT default model with 52.5% fewer hallucinated claims on high-stakes queries than GPT-5.3 Instant. (TechCrunch)
Perplexity Computer (May 4): Perplexity launched a Mac app integrating local files, apps, and sessions for Pro and Max subscribers, with access to CB Insights, PitchBook, and Statista data. (Perplexity changelog)
BrandCited's structured data audit checks FAQPage schema as part of its AEO (Answer Engine Optimization) score. If your pages are missing FAQ schema, the audit shows a critical or warning finding with an estimated citation impact score. If your FAQ answers are too long, contain promotional language, or use inconsistent entity naming, the audit flags those quality issues separate from the presence check.
Run the full check free at brandcited.ai. Results show your AEO score alongside specific structured data gaps, ranked by estimated citation impact across each of the 9 AI platforms BrandCited monitors.
1Keep your existing FAQ schema. The rich result feature is gone. The citation signal is not.
2Add FAQ schema to your homepage. If your homepage has no FAQPage markup, that is the highest-priority gap. Add 4-5 questions that real users type into AI assistants about your product category.
3Write answers for AI extraction. Each answer should be complete in 40-60 words, self-contained, and name your brand by name, not "we" or "our product."
4Add schema to product and pricing pages. These pages appear in commercial intent queries. Without FAQ schema, you are invisible in AI-generated purchase recommendations.
6Run a BrandCited scan. The audit shows which pages are missing FAQ schema and ranks the fixes by estimated impact across all 9 AI platforms.
Run a free AI visibility audit on your brand at brandcited.ai. You'll see your score across 9 AI platforms in 30 seconds, with every structured data issue ranked by impact.
Does FAQ schema still work after Google deprecated FAQ rich results?
Yes. Google deprecated the visual FAQ rich result on May 7, 2026, not the FAQPage schema type. Google's documentation states it will continue using FAQ structured data to understand pages. Every major AI platform, including ChatGPT, Perplexity, Gemini, and Google AI Overviews, continues to parse FAQPage markup as a citation signal. Pages with FAQ schema are 3.2x more likely to appear in AI Overviews.
Should I remove FAQ schema from my website?
No. Removing FAQPage schema removes a signal that AI platforms use to identify and cite your content. Pages with FAQ schema achieve a 67% citation rate in AI responses for relevant queries. Keeping the schema costs nothing. Removing it costs citations across ChatGPT, Perplexity, Gemini, and Google AI Overviews.
FAQ schema vs. organic SEO rankings: which drives more AI citations?
They are not interchangeable. Only 38% of AI Overview citations come from pages ranking in the top 10, according to ALM Corp analysis. A page with strong FAQ schema can earn AI citations from a much lower organic rank. Both factors contribute, but FAQ schema is the more direct signal for AI citation.
How do I write FAQ answers that get cited in AI search?
Write each answer as a standalone, complete response to the question, with no setup or reference to other page sections. Keep answers under 60 words. Name your brand by name, not "we" or "our platform." Include one specific number or verifiable fact per answer when you have it. AI models quote self-contained factual sentences more often than narrative paragraphs.
How do I check whether my FAQ schema is getting me AI citations?
Run a BrandCited scan at brandcited.ai. BrandCited's AEO audit checks FAQPage schema across your site, scores the quality of your question-answer pairs, and shows which of the 9 AI platforms are citing your FAQ-marked content. The scan takes 30 seconds and covers ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews.
Track your AI visibility now
See exactly how AI engines like ChatGPT, Perplexity, and Gemini perceive your brand.
The BrandCited team covers GEO, AI search optimization, and brand visibility strategy. We publish research, practical guides, and product updates every week.
