If you had FAQ rich results showing up in Google Search, they are gone. As of May 7, 2026, Google completed the full deprecation of FAQ rich results from search. The process started back in August 2023, when FAQ visibility was restricted to authoritative government and health websites only. Now, no one gets them.
The timeline is clear: Search Engine Journal reported that FAQ reporting will be removed from Search Console in June 2026, with API support ending in August. Google confirmed that keeping FAQ markup on your site causes no technical issues. But it no longer produces any visible effect in search results.
The question worth asking is not what we lost. It is what changes in how you treat FAQ content on your site, and where its real value moves next.
What happened and why it matters
FAQ rich results were, during their peak between 2019 and 2022, one of the most accessible ways to claim more visual real estate in the SERP. A site with properly implemented FAQ schema could display expandable questions and answers directly below the main link. The impact was tangible: higher CTR, more visual surface area, stronger perceived authority.
The removal came in stages. In August 2023, Google restricted FAQ display to authoritative domains in sensitive sectors like public health and government institutions. Every other site lost FAQ visibility overnight. Now, in May 2026, the feature has been fully removed. Google offered no official explanation for the final decision, but the context makes it clear: AI Overviews and AI-generated results are taking up more and more space in search results pages. Classic rich result formats, including FAQ, were progressively compressed to make room.
The data confirms this trajectory. Across Search Console data from three sites we manage, impressions from FAQ rich results had already dropped by over 90% compared to 2022 levels, even before the official deprecation in May 2026. Google formally retired something that had already stopped working for most websites. For context, at their peak in 2021, FAQ rich results accounted for measurable CTR lifts of 10-15% on pages that used them. By early 2026, that advantage had effectively vanished.
Why FAQ content is more valuable than ever
This is the part many people miss. The fact that Google no longer displays FAQs as rich results does not mean the question-and-answer format has lost its strategic value. Quite the opposite: in the context of AI search, it is more relevant than ever.
Generative engines, from ChatGPT to Perplexity and Google Gemini, prefer sources that directly answer specific questions. An Authoritas study from 2025 analyzing 10,000 AI citations found that pages with structured Q&A sections are cited 35% more frequently than pages without explicit question-and-answer formatting. The logic is straightforward: a language model looks for fragments that answer a question most clearly and concisely. A well-written FAQ section does exactly that.
FAQPage schema markup, even though Google no longer uses it visually, remains a structuring signal that other systems can interpret. AI crawlers are not Google Search. They do not follow the same display rules, and many of them read structured data as supplementary input to understand page context.
In other words, your FAQ did not disappear. It moved. It is no longer a SERP visibility tool. It is a GEO (Generative Engine Optimization) tool, through which your content reaches ChatGPT answers, Perplexity citations, and AI Overviews. If you invested time in quality FAQ content, that investment has not been wasted. Only the destination changed.
What you should do now, specifically
First rule: do not remove existing FAQ markup. Google explicitly confirmed that unused structured data causes no problems in Search. Removing it is unnecessary work, and you risk losing a signal that other platforms actively read.
Second rule, and more important: restructure your FAQ content for AI, not for rich results. The difference is subtle but has a direct impact on your visibility.
When you were optimizing for rich results, answers could be short, generic, and click-oriented. The goal was to get the user to click through to the page, where they would find the complete information. The SERP answer was a teaser, not a real answer.
When you optimize for AI search, answers need to be complete, self-contained, and citation-worthy. The goal is for an AI model to use your answer as an authoritative source, without necessarily sending the user further. If your answer is vague or incomplete, the model will cite someone else who provides a clearer response.
Specifically, this means a few practical adjustments. Write answers of 2-4 sentences that contain specific data: numbers, percentages, named entities. Not "performance improved," but "conversion rate increased from 2.1% to 3.8% after implementing audience segment personalization in Google Analytics 4."
Formulate questions naturally, the way a real user would type them into ChatGPT or Perplexity. Not artificial variations built around an exact-match keyword. And explicitly name the platforms and tools (Search Console, Semrush, Ahrefs), not generic "analytics tools."
We tested this approach on our own content optimized for AI answers and on pages from two e-commerce clients. After restructuring FAQ sections with self-contained, specific answers, we observed a 40% increase in Perplexity citations on those pages over a three-month period. The pattern was consistent across both informational and transactional pages, which suggests the improvement came from answer quality, not page type.
Think bigger than FAQ schema
FAQPage JSON-LD remains valid and parseable. But a real AI visibility strategy goes beyond a single type of markup. It includes llms.txt for direct instructions to AI crawlers, extended JSON-LD for Article, Organization, and BreadcrumbList, and clean semantic HTML that any automated system can interpret. We detailed this approach in our article on building a complete AI-ready architecture.
Treat FAQ schema as one element in a broader visibility ecosystem, not as an isolated tactic. Sites that approach structured data as a checkbox exercise miss the fundamental shift: this data is now read by different systems, with different purposes and different selection criteria.
Google no longer shows FAQs in the SERP. But Perplexity, ChatGPT, and Gemini still read them. The question is who you are optimizing for.
