FAQ schema in 2026: still worth it after the deprecation?

Yes, but the goal has changed. On 7 May 2026 Google retired FAQ rich results from Search, those expandable question dropdowns under a listing. The FAQPage markup is still valid and does not hurt rankings. The value has moved elsewhere: AI answer engines cite question-and-answer content at some of the highest rates of any format.

For years, FAQ schema was a way to grab more space on Google’s results page. That era is over. This guide explains what changed, what to do with the schema you already have, and why the question-and-answer format is still one of your best bets, now for AI-generated answers.

What changed with FAQ schema in 2026?

On 7 May 2026, Google deprecated FAQ rich results in Search. The accordion of questions and answers that used to expand your listing in the SERP no longer appears. It was not a surprise: back in August 2023 Google had already limited these results to authoritative health and government sites, and in May 2026 it retired them completely (Google Search Central).

Did Google remove the schema or just the rich result?

Only the rich result. This is the point most people get wrong. What disappeared is the SERP display feature, not the markup. FAQPage is still a valid Schema.org type and can stay on your pages with no problem. Google deprecated a way of appearing, not the code itself.

Does the deprecation affect my rankings?

No. Google confirmed that rankings are not affected by the deprecation. It is a search-appearance change, not an algorithmic one. Pages that previously gained extra space with the FAQ accordion may see their click-through rate shift, but organic position is not adjusted because of this.

What happens to Search Console and the Rich Results Test?

The wind-down is staged. Search Console reporting and Rich Results Test support end in June 2026, and Search Console API support ends in August 2026. If you still see FAQ data in those reports, it is temporary. It is not worth investing time to fix FAQ warnings in those tools.

Is FAQ schema still useful for AI search?

It is, perhaps more than ever. While Google pulled back, AI answer engines started extracting question-and-answer content aggressively. Research in 2026 puts FAQ content among the formats with the highest citation rates in AI answers, showing up in ChatGPT, Perplexity and AI Overviews above unstructured content (Frase, 2026).

There is an important detail about how this works. Language models tokenize JSON-LD as plain text; they do not parse the markup as structured data. In other words, ChatGPT and Perplexity do not really read your FAQPage: they extract the visible Q&A on the page, the one the schema is meant to mirror. The practical takeaway: what counts is having the questions and answers visible in the content, with the schema as reinforcement.

Should I remove FAQPage from my pages?

No. Keep the schema. It does not hurt rankings, it is still valid on Schema.org, and it helps Google’s Knowledge Graph pipeline understand the structure of your answers. Removing it is work with no upside. The only mistake would be to depend on it for SERP space, because that option is gone.

How should you use FAQ schema in 2026?

The strategy is dual: schema for machines, visible Q&A for people and for AI. Follow these rules:

• Write real questions as visible H2 or H3 headers, not only inside the schema.
• Match the question text in the schema to the visible header, word for word.
• Keep each answer between 40 and 60 words, the block AI extracts and cites most.
• Keep FAQPage JSON-LD as reinforcement, without depending on it to appear on Google.
• Use questions people actually ask, drawn from “People Also Ask” and conversational queries.

Next step: dig into the AEO pillar to master the citable answer format, and see how to apply all of this in practice in the how to rank on ChatGPT guide.