FAQ schema

Is FAQ schema still worth adding?

FAQPage schema wraps each question and its answer in a structure engines can extract whole. Add it to any page where you genuinely answer customer questions, keeping the markup to the questions and answers that are actually visible on the page, and unique per page so engines get clean pairs.

There is a wrinkle worth knowing: Google has deprioritized FAQ rich results in its own search results for most non-authority sites. But that is a SERP display decision, not a signal decision. AI engines like ChatGPT and Perplexity still lean on FAQ structure heavily when they decide what to quote, so the markup is very much worth shipping even where Google no longer shows the rich snippet.

The common failure modes are marking up questions that do not appear on the page, or repeating the same FAQ block across many pages, both of which can trigger structured-data warnings. GrowthManager generates FAQPage schema that mirrors the visible content one-to-one and keeps it unique per page.

Keep the answer text inside FAQPage schema between roughly 40 and 120 words. Shorter than that and the engine treats the answer as too thin to extract confidently. Longer and the engine starts truncating, often at sentence boundaries that change the meaning. Inside that band, write each answer to stand alone outside the page, since that is exactly how it may end up appearing in an AI chat or a featured snippet. Names, numbers, and qualifiers belong in the answer itself, not in the surrounding prose, so the extracted snippet is still accurate when read in isolation.