Does Schema Markup Actually Get You Cited by AI? What Google Says vs What Happens
Google says AI Overviews need no special schema markup. AI Mode still uses schema heavily to verify claims and trust sources. Both are true at once, here's how.
Google states there's no special structured-data requirement to appear in AI Overviews beyond normal search eligibility. But Gemini-powered AI Mode still uses schema markup to verify claims, establish entity relationships, and judge source credibility during synthesis, meaning it's not required for eligibility but does affect whether you're trusted enough to cite. Also worth knowing: Google deprecated FAQ rich results from Google Search as of May 2026.
This is one of the more confusing corners of AEO advice, because two things that sound contradictory are both accurate at the same time.
What Google actually says
Google has been explicit: there are no special extra technical requirements for AI Overviews or AI Mode beyond being indexed and eligible to appear in regular Search with a snippet. No special schema.org markup is required, and Google has said sites shouldn't overinvest in structured data purely for AI visibility.
Taken alone, that reads like "schema doesn't matter for AI search." That's not quite it.
What schema actually does for AI systems
Structured data isn't a ranking requirement, it's a trust and disambiguation layer. Gemini-powered AI Mode uses schema markup to verify claims, establish entity relationships, and assess source credibility while it's synthesizing an answer. When a model can clearly verify who you are, what you offer, and how you relate to the topic at hand, it's more willing to cite you, not because the schema itself is a ranking signal, but because it reduces the model's uncertainty about trusting the page at all.
That distinction, eligibility versus trust, is the whole answer to "does schema matter." It doesn't get you into the candidate pool. It affects whether you survive the pool's trust filter.
The schema types actually worth your time
Not all schema pulls equal weight. Five types are the ones that consistently show up as decisive:
| Schema type | Why it matters for AI citation |
|---|---|
| Organization | Establishes who you are as an entity; prioritize this first |
| Article / BlogPosting | Names author, date, publisher for content trust signals |
| FAQPage | Lets a model quote a Q&A pair directly (see caveat below) |
| Product | Structured price, availability, spec data for commerce queries |
| LocalBusiness | Location and service-area entity data for local queries |
JSON-LD is the preferred format across all of these, it's easier to manage and less prone to breaking than microdata or RDFa embedded directly in HTML.
The FAQ schema caveat
Here's the part that catches people off guard: Google deprecated FAQ rich results from Google Search as of May 2026, along with the FAQ search-appearance feature and Rich Results Test support for it. That's a Google Search rich-results deprecation specifically, not a statement that FAQPage schema is worthless for AI citation generally: it can still help a model directly quote a Q&A pair instead of extracting one from prose, which is exactly what this site's own Keywords tab checks for on a Full AEO Audit. But if you were adding FAQ schema purely to chase the old Google Search FAQ rich snippet, that specific payoff is gone.
What to actually do
- Prioritize Organization schema first. It's the foundation every other entity relationship gets built on.
- Add Article/BlogPosting schema with a real author and real dates, not placeholder values, since that's what feeds the trust-verification step directly.
- Keep FAQPage schema if you have genuine FAQ content, but don't add it expecting the old Google rich-result payoff. Add it because a model can quote it directly, which a free LLM Access Check will show you whether your JSON-LD is even reaching the crawlers that would use it.
- Don't treat "no schema required for eligibility" as "schema doesn't matter." It's a trust signal at the synthesis stage, not a gate at the eligibility stage. For the full picture of what AI crawlers can and can't see on your page in the first place, see What Is AEO?.
Sources: Opace, Schema Markup for SEO 2026; Digital Applied, Schema Markup After March 2026; Stackmatix, Optimizing FAQ Schema for Google AI Overviews.
FAQ
Do I need schema markup to appear in Google AI Overviews?
No. Google has stated there's no special structured-data requirement beyond normal Search eligibility. Schema still plays a supporting trust role once a model is deciding whether to cite you, but it isn't a gate to being considered.
Is FAQ schema still worth adding?
Yes for AI citation purposes, since it lets a model quote a Q&A pair directly. But Google deprecated the FAQ rich-result appearance in Google Search itself as of May 2026, so don't add it expecting that specific search-results payoff.
Which schema type should I add first if I can only do one?
Organization schema. It establishes the core entity everything else, Article, Product, FAQPage, gets related back to.