Structured data.
Structured data is markup that explains to search engines and AI what your page is actually about: product, article, organization, event, FAQ. What makes the difference between showing up as a link and being cited as an answer.
By Hans Sandblom · Senior SEO Specialist· Published · Updated Our view
Semantic clarity makes you citable. Guessing doesn't get you into a ranked source.
What structured data is.
Structured data is code you add to your page to explain the content semantically: this is an organization, this is an article, this is an FAQ. Without the markup the search engine only sees text and has to guess the meaning. With the markup it knows exactly what is what, and can treat it accordingly.
The dominant format is JSON-LD, a lightweight JavaScript format placed in a script tag in the page head. The vocabulary comes from Schema.org↗, a collaboration between Google, Bing, Yahoo and Yandex. Types and their fields are defined there: Product, Article, Organization, FAQPage, BreadcrumbList and so on. Alternative formats exist (Microdata, RDFa) but Google recommends JSON-LD↗.
The effect isn't direct ranking. Schema markup doesn't lift a page from position ten to five just because the markup is there. Two things happen instead: Google can show rich results (stars, prices, breadcrumbs in SERP), and AI answers get structure to cite. Both translate to measurable value through CTR and AI inclusion.
Common schema types.
Schema.org has hundreds of types, but in practice a handful recur. The ones we see most:
Organization
Your company: name, logo, contact details, social profiles. Foundation for the knowledge graph and brand identification. Belongs on the homepage or in a global site-wide script tag.
Article
Blog posts, articles, pillar content. Fields like headline, author, datePublished, mainEntityOfPage. Produces rich results for articles (author, date, image). 'Top stories', though, applies to news publishers in Google News.
Product
E-commerce products: price, stock, reviews. Produces rich results with stars and price directly in SERP. Large effect on CTR for e-commerce.
FAQPage
Frequently asked questions and answers on the page. Google restricted FAQ snippets in SERP to authoritative government and health sites in 2023, so for most sites the value now lies in AI answers citing FAQ-marked pages when the question is definitional.
BreadcrumbList
The page's place in the navigation hierarchy. Gives a clearer SERP snippet (breadcrumbs instead of URL). Small but stable CTR lift.
Service
Service pages. Defines what you offer, to whom (areaServed), and connects to the Organization entity. Helps AI answers understand your service offering semantically.
Structured data and AI search.
Classic SEO discussion treats structured data as a rich-results tool. That's true, but not the whole picture in 2026. More important is how schema markup affects inclusion in AI answers.
When an AI like AI Overview, ChatGPT or Perplexity is about to answer a question, it looks for structured sources to cite. A page that clearly declares 'this is a definition', 'this is an FAQ', 'this is an article authored by X' is both easier to parse and easier to trust for an AI. Schema becomes an entry ticket.
We follow which sources AI answers cite and what they have in common. The principle is clear: schema markup makes it easier for an AI to parse a page semantically, which is a prerequisite for being cited correctly. Schema is an entry ticket rather than a bonus. That's why we argue for it as an AEO foundation, not just a rich-results optimization.
Example from our own site: memorise.se publishes Organization, WebSite, WebPage, Article, FAQPage, BreadcrumbList and Service per relevant page as a connected @graph. Not because it's trendy but because it's the foundation for being cited by the same AI systems we study.
Common misconceptions.
Structured data is an area where expectations often land wrong. Three recurring misconceptions:
- 'Schema lifts ranking directly.' No. Google has repeatedly explained that schema in itself isn't a ranking factor. It enables rich results (which lift CTR) and helps AI cite. The ranking lift is indirect, via CTR and AI inclusion.
- 'More schema is better.' No. Mark up what is relevant and true for the page. Placing Product schema on a blog post, or oversized nested schema, creates validation errors and can trigger manual action. Less and correct beats more and careless.
- 'Google Rich Results Test↗ validates = everything works.' Partly. Rich Results Test catches syntax errors and which rich results can trigger. It doesn't catch semantic wrongness (declaring you're company X when you're company Y) or knowledge graph linkage. There's more than one step in validation.
How we work with structured data.
We build schema when it delivers measurable value. All sites need Organization and BreadcrumbList. E-commerce needs Product with price and stock. Knowledge sites (like ours) need Article and FAQPage. Service sites need Service. We prioritise by what actually triggers rich results or AI inclusion for your specific industry.
We implement in-house when we have platform access, or deliver as templates to your dev team. All schema we build is validated against Google Rich Results Test and the Schema.org Validator. You own the code, no black box.
For AEO engagements, structured data is usually the starting point. A site without schema in place is hard to build citation volume on, and schema is often the effort that produces the fastest measurable effect on AI inclusion.
Get a free schema audit.
Send us your URL and we'll review current schema markup, validate, and come back with a concrete action list.
Write to us →Frequently asked questions about structured data
Is structured data good for SEO?
Yes, but not as a direct ranking factor. Schema markup enables rich results (stars, prices, breadcrumbs in SERP), which lifts CTR. It also raises the chance that AI answers cite the page. The effect is indirect but measurable.
What is structured data?
Code placed on the page to explain the content semantically to search engines and AI. The dominant format is JSON-LD, with vocabulary from schema.org. It says: this is an article, this is a product, this is an FAQ.
What's an example of structured data in SEO?
An e-commerce page marking Product schema with price, stock and review average. Google can then show 4.7 out of 5 stars and 499 kr directly in search results instead of just a URL. Tangible CTR effect.
What's the difference between JSON-LD and Microdata?
Both are formats for structured data. JSON-LD is a script block in the page head, separated from HTML. Microdata is inlined in HTML tags. Google recommends JSON-LD because it's easier to maintain and validate. The effect is the same when both are correct.
How does structured data affect AI search?
AI answers (AI Overview, ChatGPT, Perplexity) look for structured, clearly declared sources to cite. A page with FAQPage schema on an FAQ, Article schema on an article, is easier to parse correctly. Schema becomes a prerequisite for the source being lifted as an answer, not just as a link.