Structured Data for AI Search: What It Does and What It Doesn't

Structured data is code that tells search engines, in a standard language, what the information on your page is: a product, an article, a business or an event. Google's 2026 documentation is clear: you do not need structured data to appear in AI Overviews and AI Mode, and there is no special schema.org markup you need to add. Structured data is still worth having, but its value lies in rich result eligibility and in letting machines read your page's information without errors, not in a promise that it will make you stand out in AI.

What is structured data?

Structured data labels the elements on a page using a shared vocabulary called Schema.org. On a product page, you state separately which item is the product name, the price, the currency, the stock status and the rating; on a blog post, which is the headline, the author and the publication date.

Google supports three formats: JSON-LD, Microdata and RDFa. Google recommends JSON-LD if your site's setup allows it, because it is added as a separate script block without touching the page's visible HTML, which makes it the easiest method to manage on large sites. Most content management systems and e-commerce platforms generate basic markup in JSON-LD on their own.

Does Google require schema for its AI features?

No. Google's "AI features and your website" documentation states plainly that you do not need to create new machine-readable files, AI text files or markup to appear in these features, and that there is no special schema.org structured data you need to add. The guide to optimizing for generative AI features, published in May 2026, follows the same line: structured data is not required for generative AI search. The same guide says that llms.txt files are ignored by Google Search and that you do not need to break content into small chunks for AI.

So be cautious about services sold as "add schema and ChatGPT or Google AI will cite you". Google's generative AI features rely on its core ranking and quality systems. What determines visibility is still whether the page can be crawled, whether it answers the question clearly, and whether it offers information not found elsewhere. As for how other AI assistants use structured data, detailed public documentation is limited, so promising a definite benefit there would not be honest.

Why does structured data still matter?

The same Google guide recommends continuing to use structured data as part of your overall SEO strategy. The concrete benefits are:

• Rich result eligibility: Displays such as products, review stars, events, recipes, videos and job postings do not appear without structured data. There is no guarantee, but without markup there is no eligibility either.
• Removing ambiguity: Text alone does not always make clear whether a number on the page is a price or a model number. Markup says so explicitly.
• Consistent organization details: Organization markup defines your business name, logo, contact details and official social profiles in one place.
• Product data: For e-commerce, Google recommends Merchant Center product feeds, and Google Business Profile for businesses, so that products can appear in search results, including AI responses. Keeping the Product markup on the page identical to the information in the feed strengthens consistency.

What is FAQ schema good for now?

For a long time the advice was "add FAQ schema so your questions and answers expand in search results". That no longer holds: Google announced that FAQ rich results are no longer shown in Search from 7 May 2026, and removed the documentation for this feature in June 2026. FAQ no longer appears in the current structured data gallery.

Existing FAQPage markup on your pages does no harm, and you do not need to rush to remove it. But marking up a new FAQ section in the hope of a rich result no longer has any visible effect on Google. The value of an FAQ section now goes directly to the reader: answering frequently asked questions clearly on the page presents the information plainly to visitors and to any system that reads the page. We look at how to write those answers in our article on AEO and FAQs.

Which markup goes on which page?

As of 2026, the structured data gallery Google supports includes types such as Article, Breadcrumb, Event, Local business, Organization, Product, Profile page, Q&A, Recipe, Review snippet, Software app and Video. These are the pairings that most often help on small and mid-sized business sites:

• Home and About pages: Organization (LocalBusiness if you receive customers at a physical location).
• Product pages: Product; Review snippet if you have genuine customer reviews.
• Blog posts: Article.
• All inner pages: BreadcrumbList, which shows where the page sits within the site.
• Event and training announcements: Event.
• Pages with video: VideoObject.

Adding every possible type to every page is not the goal. Choose the type that describes what the page really is. Google's guidelines give the example of a site marking up woodworking instructions as a recipe as irrelevant markup; marking up a service page as a product is the same kind of mistake.

How do you implement it correctly?

1. Identify your key page templates. Think in templates rather than individual pages: the product template, the blog template, the contact page.
2. Mark up visible content. Under Google's general guidelines, content the reader cannot see should not be marked up; the information described in JSON-LD should also be in the body of the page. Marking up a rating or price that is not on the page can lead to loss of rich result eligibility and a manual action.
3. Complete the required properties. According to Google, items missing required properties are not eligible for rich results. Less information that is accurate and complete beats more information that is careless.
4. Test before publishing. Use the Rich Results Test during development and the Schema Markup Validator to check syntax.
5. Monitor after publishing. The rich result status reports in Search Console show errors, and the URL Inspection tool lets you check whether Google found the markup.
6. Keep dynamic information in sync. Fields such as price and stock should be fed from the same source on the page, in the markup and in the product feed. Hand-written JSON-LD is where things break most often when a price changes.

How do you measure the impact?

For rich results, use the enhancement reports in Search Console and the search appearance filters in the Performance report. On the AI side, since 31 August 2026 all sites have a "Generative AI performance report" that shows your impressions in AI Overviews and AI Mode but no click data. It is hard to attribute changes in that report to structured data, because Google does not treat markup as a requirement for these features. When you measure, do not assume markup is the only variable; also note any content changes you made in the same period.

If you want to handle structured data as part of your SEO work, take a look at our SEO service or our approach to AI search optimization.