Structured data is a standard way to mark up the information on a page so search engines and AI systems can read it as machine-readable facts. The most common format is JSON-LD code that uses the Schema.org vocabulary.
A person reads “Stradiji is an SEO and GEO consultancy founded in Istanbul in 2010” and gets it. A machine has to guess which word is the company name, which is the city and which is the founding year. Structured data removes the guesswork: “This is an Organization. Here’s its name, its founding date and its address.”
Why structured data matters for your brand
For years, the main payoff of structured data was rich results: star ratings, FAQ drop-downs, recipe cards. That payoff has shrunk. In August 2023, Google limited FAQ rich results to well-known government and health sites (Google Search Central, August 2023). In June 2025, it announced it would phase out seven more rich result types, including Course Info, Claim Review and Estimated Salary (Search Engine Land, June 2025).
But the real job of structured data isn’t decoration. It’s meaning. At SMX Munich in March 2025, Microsoft’s Fabrice Canel confirmed that schema markup helps Microsoft’s large language models understand content (Search Engine Land, March 2025). Rich results may go, but an engine that understands you correctly stays.
How structured data works
- Vocabulary. Schema.org defines shared types such as Organization, Person, Article, Product and FAQPage, plus the properties of each type.
- Format. These types usually go on the page as JSON-LD, in a code block separate from the visible content.
- Connections. Properties such as sameAs, author, publisher and @id link entities together: “This article’s author is this person. This person founded this company. These are the company’s official profiles.”
- Validation. Engines check whether the markup matches what’s visible on the page. Marking up information people can’t see erodes trust.
The SEO and GEO connection
Structured data is the bridge between SEO’s technical foundation and GEO’s entity layer. On the SEO side, it helps engines classify your page correctly. On the GEO side, it helps models recognize your brand as the right entity. Organization schema and sameAs links are also where entity optimization starts.
One caution: schema alone doesn’t earn citations. If the information you mark up is missing from the page, thin or contradicted by outside sources, code won’t fix it. Schema makes good content readable. It doesn’t make content good.
What we’ve seen in the field
About 30% of the more than 100 websites we audited through September 2026 lacked even basic Organization schema (The Three Doors of AI Visibility). The mistakes we see most:
- The same schema added twice. In October 2026, we found that FAQ schema we’d added by hand on our own site duplicated the FAQ schema the site already generated automatically, and we removed one. Check your plugins, theme and hand-coded markup together before you add schema.
- Markup that doesn’t match the visible content: ratings, prices or questions that aren’t on the page.
- Empty sameAs fields, or sameAs links pointing to old, closed profiles.
A practical structured data playbook
- Start with Organization. Name, logo, founding date, founder, address, contact details and sameAs links to every official profile.
- Connect your authors. Article schema on articles, Person schema for the author and a link to the author page.
- Match schema to page type. Service on service pages, Product on product pages, FAQPage on FAQ sections. Don’t put every type on every page.
- Keep it identical to what’s visible. Every fact you mark up should be readable on the page.
- Remove duplicates. Check whether your theme, SEO plugin and hand-coded markup produce the same type twice.
- Validate regularly. Use Google’s Rich Results Test and the Schema.org validator after every major site change.
Related terms
Frequently asked questions
Does structured data directly improve rankings?
No. It isn’t a direct ranking factor. It helps engines understand your page and your brand, and it can affect clicks by making you eligible for rich results.
Is FAQPage schema still worth it now that FAQ rich results are gone?
For most sites, Google no longer shows the visual FAQ rich result. But describing your Q&A in machine-readable form still helps engines understand the content. What matters is that the questions in the schema appear word for word on the page, and that there’s no duplicate.
Which schema type should I start with?
Organization. Once your brand name, logo, company details and official profiles are defined in one place, other schema (Article, Person, Service) can connect to that entity.
Do AI models read schema?
Microsoft confirmed in March 2025 that schema helps its large language models understand content. Other engines haven’t said so as clearly. But schema feeds the understanding layer of the search indexes those engines already rely on.
