Powered by KwikChexStructured data for AI search: what helps and what doesn't
By About This Business editorial team · Published · Updated
Short answer
Structured data is machine-readable markup, usually schema.org in JSON-LD, that describes what a page is about. It helps search systems read a page accurately, but Google says no special markup is needed for its AI features and that structured data must match the visible text. For a business, Organization, Article, FAQPage and BreadcrumbList are the useful basics.
Key takeaways
- Structured data describes a page for machines. It doesn't replace clear visible text.
- Google says no special schema is required for AI Overviews or AI Mode.
- Markup must match what's on the page. Claims that appear only in markup are a mistake.
- Useful basics: Organization, Article, FAQPage and BreadcrumbList.
What is structured data?
Structured data is machine-readable markup that labels what a page contains, such as "this is an organisation", "this is an article" or "these are questions and answers". It's usually written in schema.org vocabulary using JSON-LD, a block of code in the page that visitors don't see. It's one part of How AI assistants decide what to say about your business.
Do AI features need special structured data?
No. Google says there are no additional requirements for AI Overviews or AI Mode, and no special schema.org markup is needed. Google also says you don't need new machine-readable files or AI-specific text files. Structured data helps systems interpret a page, and it should match the visible text.
Which structured data types suit a business site?
Four types cover most needs.
| Type | Use it on | What it describes |
|---|---|---|
| Organization (or LocalBusiness) | Home or about page | Name, logo, website, official profiles |
| Article | Guides and posts | Headline, author, publisher, published and modified dates, image |
| FAQPage | Pages with a visible FAQ | Each question and its answer, exactly as shown |
| BreadcrumbList | Every inner page | Where the page sits in the site's structure |
What's the golden rule?
Markup must describe what's actually on the page, and nothing more. If your FAQ markup contains an answer the page doesn't show, or your Organization markup claims an accreditation your page doesn't mention, the markup is misleading. For Business DNA, that rule goes further: structured data carries the same per-fact sources and statuses as the visible profile, so a company-provided fact is never presented to machines as authenticated. See Company-provided vs authenticated.
What matters more than markup?
Crawlable, server-rendered text, clear headings and internal links.
- Text in HTML. Key facts should be readable without running scripts, opening PDFs or reading images.
- Answer-first sections. Each section opens with a direct answer that makes sense on its own.
- Internal links. Google lists internal links as a basic way to make content easy to find. A pillar page linking to related articles, and back, shows how topics connect.
- Visible dates and authors. Readers and systems can both see how current a page is.
How do you check your structured data?
Validate it, then compare it against the visible page.
- Test the page with Google's Rich Results Test or the Schema.org validator.
- Read the JSON-LD and check every claim appears on the page.
- Check dates, names and addresses match your other listings. See When AI gets your business wrong.
Thinking of adding an llms.txt file too? See llms.txt: what it is and whether you need it.
Part of How AI assistants decide what to say about your business.
FAQ
Will adding schema markup get me into AI answers?
Not on its own. It helps systems read your page accurately, but the page still has to be crawlable, relevant and clear. Google says no special markup is required for its AI features.
Can I put facts in structured data that aren't on the page?
You shouldn't. Google says structured data should match the visible text. Hidden claims can mislead users and systems, and they undermine trust in the page.
Which format should I use?
JSON-LD is the most common format for schema.org markup and is easy to keep separate from your page layout.