Structured data is a standard way to describe a page to a machine: what it is, who wrote it, what it sells. Our glossary has the plain definition. This guide is about whether it matters for AI answers.
Short version: it is worth doing correctly for the features Google supports, and it is not a switch for AI citations. The engines say little else.
What does Google say about schema and AI features?
Google's AI features documentation says you do not need to create new machine-readable files, AI text files or markup to appear in AI Overviews or AI Mode, and that there is no special schema.org structured data to add. It says structured data should match the visible text on the page.
Google's May 2025 post says to use structured data to describe content in a machine-readable way, to follow Google's guidelines, to make sure all the content in your markup is also visible on the page, and to validate it.
What does Google promise about structured data?
Little beyond eligibility. Google's introduction to structured data describes it as a standardized format for giving information about a page. It recommends JSON-LD, though microdata and RDFa are also supported. An object must include every required property to be eligible for enhanced display, and Google says it is more important to supply fewer but complete and accurate recommended properties than many incomplete ones. It makes no promise that a rich result will appear.
What do the other engines say?
We read OpenAI's crawler documentation and Perplexity's bots guide and help page, and Microsoft's June 2026 Bing Webmaster Tools post, on 12 November 2026. None of them says that structured data affects whether a page is cited in an AI answer, and none mentions it as a requirement. That is not proof that markup has no effect on those systems. It means no engine we read documents one, so any claim of an effect is an inference, not a published rule.
Which markup still has a job?
Markup earns its place where Google supports a feature for it. Google's search gallery lists 25 feature types at the time we read it, including article, breadcrumb, organization, product, review snippet, local business, video, event, job posting and software app. Match the type to the page: article markup on an article, product markup on a product page, breadcrumb markup where breadcrumbs show.
Even where no rich result appears, accurate markup is a clear, machine-readable description of what a page is and who stands behind it. That is useful, low in risk and consistent with how search engines already work.
What about FAQ markup?
Do not add it for rich results. Google's page on FAQ structured data says the FAQ rich result is no longer shown in Search, from 7 May 2026. Questions and answers still belong on the page for readers, and the visible text is what any engine reads first. If the old markup is already in place and accurate, it does no harm, but it earns nothing there.
How do you add structured data safely?
- Choose the type that fits the page, from Google's gallery, and fill every required property.
- Use JSON-LD unless your platform makes another format easier to maintain. Google says all three formats are fine if valid.
- Match the page. Every fact in the markup, such as price, hours, author and rating, must be visible on the page and true.
- Test it in Google's Rich Results Test and check Search Console's reports for errors.
- Keep it current. When a price, address or author changes, change the markup in the same edit.
- Never mark up what is not there, such as reviews you do not display.
What should you not do?
Do not stuff pages with every type you can find, add markup describing content that is not on the page, mark up reviews you wrote yourself, or pay for a promise that schema will win AI citations. Google says markup should describe what is visible on the page. Also avoid spending effort on markup while a page is slow, thin or blocked: the basics in our guide on how to optimize for AI search come first. The structured data entry in our glossary has the plain definition.
Conclusion
Schema is worth doing correctly for the features Google supports and as an honest description of your pages. It is not a documented route to AI citations. Match it to the page, test it, keep it current and put clear writing first.
PromotEdge International is an international strategy, technology, marketing and design company, and our delivery team is in India. See our technical audits and our AI search service.
Facts we rely on: see our company facts for what we state about ourselves and where each fact comes from.