FAQ schema is the JSON-LD markup defined by schema.org/FAQPage. It tells search engines and AI answer engines that specific content on a page is a list of frequently asked questions paired with answers. The structured format makes Q&A pairs trivially extractable.
Adding FAQ schema to high-traffic pages is the single highest-ROI AEO move in 2026. AI engines extract FAQ JSON-LD verbatim, often citing the answer text directly inside ChatGPT, Claude, and Perplexity responses. Most teams see measurable citation rate lift within 30 days of shipping FAQ schema.
Best practice: write 5 to 8 Q&A pairs per page, with each answer 2 to 4 sentences long and self-contained. Avoid promotional language; AI engines downrank obvious marketing in FAQ blocks. The questions should match real buyer queries.
Read next
Frequently asked
What is FAQ schema in SEO?
FAQ schema is JSON-LD structured data (schema.org/FAQPage) that explicitly marks question-answer pairs on a page. Search engines and AI engines use it to extract Q&A directly for rich results and citations.
How many FAQs should I add per page?
5 to 8 is the sweet spot. Fewer than 3 doesn't earn rich result eligibility. More than 10 dilutes the focus per question. Each answer should be 2 to 4 sentences, self-contained.
Does FAQ schema work for AI engines, not just Google?
Yes, broadly. ChatGPT, Claude, and Perplexity all extract FAQ JSON-LD when crawling pages. Citation rates lift measurably across all major AI engines after adding FAQ schema.