Why schema matters for AI search
Schema doesn't directly cause citations — it makes your content easier to parse, attribute, and trust. LLMs prefer pages where they can clearly identify the entity, the author, the date, and the answer. Clean schema lowers the cost of citing you.
The six schemas that earn their keep
- Organization — once, in site-wide head. Include name, url, logo, sameAs (LinkedIn, Crunchbase, Clutch, G2), founder, foundingDate.
- Person — on author bios and About page. Include name, jobTitle, sameAs LinkedIn, knowsAbout.
- FAQPage — on every page with Q&A content. Single highest-leverage schema for AI Overviews and ChatGPT citations.
- Article — on blog posts and guides. Include headline, datePublished, dateModified, author, publisher.
- Service — on service pages. Include serviceType, areaServed, provider.
- LocalBusiness — once, in site-wide head. Include address, geo, openingHours, areaServed, priceRange.
Schemas to skip
Don't waste cycles on: HowTo (Google deprecated rich-result eligibility for most queries), Speakable (limited support), exotic Course/Event schema unless that's literally your business. The six above carry 95% of the value.
How to validate
Three tools: Google Rich Results Test (validates schema for Google rich results), Schema.org Validator (validates JSON-LD syntax), and live ChatGPT/Perplexity testing (the only test that matters in 2026 — ask the engine your target prompt and see if you're cited).