SEO & Marketing

Structured Data: What Google and AI Really Read

Structured data makes content unambiguous for machines. Which types Google really uses, what AI does with it and where the limits are.

of Josef KujawskiPublished: 22 October 2026Updated: 13 October 20266 min read1.131 Words

Structured Data: What Google and AI Really Read

Short answer

Structured data is machine-readable information in the source code, usually in JSON-LD format using the Schema.org vocabulary. It tells search engines unambiguously what a page is about: a company, article, product or event. Google uses it to understand content better and to show rich results for certain types, but does not guarantee them. For AI search it is a helpful signal but does not replace good, visible content. Important: the markup must match the visible content.

What people see and what machines read

A person recognises at a glance that a page is a recipe, an event or a company profile. A machine has to infer it. Structured data does that work for it: it describes in the source code exactly what the page is about. ^[1]

Diagram: web page for people and JSON-LD for machines with the same content — Structured data translates visible content for machines.

What it looks like

The common format is JSON-LD, a small data block in the source code. The vocabulary comes from Schema.org, a shared standard of the major search engines. ^[4] A simplified example for a company:

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Example Ltd",
  "url": "https://www.example.com",
  "logo": "https://www.example.com/logo.png"
}

Google supports JSON-LD, Microdata and RDFa and recommends JSON-LD because it is easiest to maintain. ^[1]

Which types are worth it

Type	For	Possible display on Google
Organization / LocalBusiness	Company, logo, contact	Knowledge panel, logo
BreadcrumbList	Page hierarchy	Path instead of URL in the result
Article / BlogPosting	Articles with author and date	Better classification in search and news
Product + Offer	Products with price and availability	Price, reviews, stock status
Event	Events	Date and location in the result
FAQPage	Questions and answers	Rich result only for selected websites

An honest note on FAQ markup: since 2023 Google only shows FAQ rich results for well-known, authoritative government and health websites. ^[3] The markup does no harm and still describes the content clearly, but it is no longer a visibility trick.

Kernaussage

Structured data is not a ranking shortcut. It makes content unambiguous and can lead to more prominent results if the content itself is convincing.

And what does AI read?

Many promises are circulating here. The documented state:

Google says structured data helps make content understandable for its systems, but requires no special markup for AI Overviews. The fundamentals are the same as for regular search. ^[5]
ChatGPT, Perplexity and others do not disclose how much they weight markup. They read pages as text, though. Information that exists only in markup and not visibly on the page is worth little to them.
Indirectly, clean markup still helps: it strengthens the unambiguous association of companies, authors and topics, and many AI systems rely on search indexes.

Our conclusion: structured data is a good foundation for GEO, but no substitute for clear, visible answers. How AI systems choose sources is described in How ChatGPT and Perplexity choose sources.

Data as a connected picture

Tree structure: Organization connected to WebSite, Article and Service — Good markup connects pages into a coherent picture.

Structured data is strongest when it is connected: the organisation publishes the website, an article has an author who belongs to the organisation, a service is offered by it. This gives machines a clear picture of who you are and what you stand for.

Structured data is like a neatly labelled warehouse. It does not make good goods better, but it makes sure they are found.

– Josef Kujawski, Managing Director Maluure GmbH

The key rules

Only mark up what is visible. Markup must reflect page content, otherwise manual actions may follow. ^[2]
Choose the most specific type, such as LocalBusiness rather than just Organization if you have a shop.
Fill in all required fields that Google requires for the type.
Test regularly with the Rich Results Test and in Search Console.
Generate automatically where possible, so markup and content do not drift apart.

On our own website we generate markup for articles, FAQs and breadcrumbs directly from the CMS content. More on the technical foundation in Technical SEO.

More examples: LocalBusiness, Article and Event

For a shop or office, LocalBusiness is usually more fitting than Organization. A simplified example:

{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "Example Ltd",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "Example Street 1",
    "postalCode": "50667",
    "addressLocality": "Cologne",
    "addressCountry": "DE"
  },
  "telephone": "+49 221 000000",
  "openingHours": "Mo-Fr 09:00-18:00"
}

For blog articles, Article or BlogPosting describes author, date and cover image:

{
  "@context": "https://schema.org",
  "@type": "BlogPosting",
  "headline": "Structured data explained",
  "datePublished": "2026-10-22",
  "author": { "@type": "Person", "name": "Jane Doe" },
  "image": "https://www.example.com/image.webp"
}

And for events, Event states date, location and organiser:

{
  "@context": "https://schema.org",
  "@type": "Event",
  "name": "Autumn workshop",
  "startDate": "2026-11-12T18:00",
  "location": { "@type": "Place", "name": "Example Ltd", "address": "Cologne" }
}

All values are placeholders. What matters is that every detail is equally visible on the page.

Rich snippets: what is realistic

Rich snippets, meaning enhanced results with stars, prices or dates, are the most visible effect of structured data. But Google decides itself whether and when they are shown. Even error-free markup does not automatically lead to them. Products with price and availability, events, recipes and breadcrumbs have the best chances. Google generally does not show review stars for your own business on your own website.

Kernaussage

Error-free markup makes you eligible for rich snippets, but not certain to get them. Google makes the decision.

Checking schema markup: the key tools

Tool	For	Note
Rich Results Test	Checks whether a page is eligible for rich results	Only shows types Google supports
Schema.org Validator	Checks markup against the full Schema.org vocabulary	Independent of Google
Google Search Console	Shows errors and warnings across the whole website	Ideal for ongoing monitoring
Browser source code	Quick check whether JSON-LD is delivered at all	Especially important for JavaScript websites

Common mistakes

Markup for invisible content: for example reviews that are not visible on the page at all.
Wrong types: a blog article marked up as a Product to get stars.
Outdated details: prices, opening hours or dates not kept up to date in the markup.
Duplicate markup: plugins and theme generate contradictory data in parallel.
Generated only in the browser: if JSON-LD is loaded via JavaScript, some crawlers may miss it. Server-side markup is more robust.

Structured data in WordPress, Webflow and Lovable

In WordPress, SEO plugins usually handle basic markup, custom types need extra configuration. Webflow allows custom code per page and dynamic markup via CMS fields. For websites developed with Lovable, markup can be generated directly from the content and delivered server-side. How we do this for our own website is described in SEO for Lovable.

The best markup is the one nobody has to look after, because it is generated automatically from the real content.

– Josef Kujawski, Managing Director Maluure GmbH

Conclusion

Structured data is a translation aid for machines. Google uses it for understanding and rich results, AI systems benefit more indirectly. Used correctly, unambiguously and honestly, it belongs to the basic equipment of every professional website.

Frequently asked questions

What is structured data?+

Structured data is standardised, machine-readable information in a page's source code. It describes unambiguously whether a page is, for example, about a company, an article or a product.

Which format does Google recommend for structured data?+

Google supports JSON-LD, Microdata and RDFa and recommends JSON-LD because it is easiest to implement and maintain.

Does structured data improve my ranking?+

Not directly. It helps Google understand content and can lead to more prominent rich results, which can increase click-through rate.

Are FAQ rich results still shown?+

Since 2023 Google only shows FAQ rich results for well-known, authoritative government and health websites. FAQ markup does no harm and still describes the content unambiguously.

Do ChatGPT and Google AI use structured data?+

Google says structured data helps make content understandable for its systems, but requires no special markup for AI Overviews. How other AI providers weight markup is not disclosed.

Which structured data does a business website need?+

At least Organization or LocalBusiness on the homepage, BreadcrumbList for navigation and, depending on page type, Article, Product, Service or Event.

How do I test structured data?+

With Google's Rich Results Test and the Schema.org Schema Markup Validator. Search Console also shows errors and warnings.

May markup contain content that is not on the page?+

No. Google requires structured data to reflect the visible content. Diverging or misleading information can lead to a manual action.

Does structured data guarantee rich snippets?+

No. Error-free markup makes a page eligible, but Google decides itself whether a rich snippet is shown.

What is the difference between the Rich Results Test and the Schema.org Validator?+

Google's test only checks types for which Google offers rich results. The Schema.org Validator checks markup against the full vocabulary, independent of Google.

Sources

Einführung in strukturierte Daten, Google Search Central (2025)
Allgemeine Richtlinien für strukturierte Daten, Google Search Central (2025)
Vereinfachung der Suchergebnisse: FAQ und HowTo, Google Search Central Blog (2023)
Schema.org Vokabular, Schema.org (2025)
KI-Funktionen und deine Website, Google Search Central (2025)

#Structured Data#Schema.org#Technical SEO#GEO

Author

Josef Kujawski

Managing Director & Creative Director at Maluure

Josef leads the strategic and creative development of brands at Maluure. For over a decade, he has guided medium-sized businesses from brand positioning to digital delivery.

Managing Director of Maluure GmbH, Cologne. Specialising in corporate design, brand strategy and digital brand experiences.

LinkedIn Website