Structured data (schema): how Google and AI understand your website
In short: structured data (schema) is code that describes your page’s content in a format machines understand: this is a company, this is its address, this is an article by so-and-so, this is a service in such-and-such a city. It helps Google understand the page and may make it eligible for rich results, and it must match exactly what visitors see. The key types for a company: Organization, WebSite, Service, Article, BreadcrumbList, and FAQPage when the questions are visible on the page. Test it with Google’s tool before publishing.
A visitor understands at a glance that your page is a service page for a company in Jeddah. A machine needs someone to tell it so explicitly. That’s the job of structured data, and its importance has grown as search and AI engines rely on understanding content, not just reading it.
What is structured data?
According to Google’s introduction to structured data, it’s a standardised format for providing information about a page and classifying its content, which Google uses to understand the content and may display more richly in results. The format Google usually recommends is JSON-LD: a block of code in the page that doesn’t change how it looks.
Which types do you need?
- Organization: company name, logo, website, contact details, official accounts, and city.
- WebSite: the site’s name and language.
- Service: for each service page: its name, description, provider and the area it serves.
- Article / BlogPosting: for articles: headline, author, publish and update dates, and image.
- BreadcrumbList: the navigation path, sometimes shown in results instead of the URL.
- FAQPage: for frequently asked questions actually visible on the page.
- Product and Offer: for stores: price, availability and ratings.
- LocalBusiness: for businesses customers visit at a specific place.
The golden rule: match the page
Google stresses on its page about AI features that structured data must match the visible text on the page. Don’t add questions that aren’t on the page, ratings customers didn’t write, or prices different from those displayed. Misleading markup can cost you rich results.
FAQs: still useful?
In 2023 Google reduced FAQ rich results to well-known, authoritative government and health websites. So you’ll usually not see the questions under your link on Google. But the FAQ section itself is still useful to visitors, gives answer engines clear text to quote, and structured data describes its structure precisely.
How do you add it?
- Ready-made platforms (Salla, Zid, and WordPress with SEO plugins) add basic types automatically; review what they add and fill the gaps.
- Custom-built sites generate JSON-LD from the page’s own data, so it always matches. (That’s how we built the MAMZ site: every service, article and city page generates its data from the same visible text.)
How do you test it?
- Google’s Rich Results Test for each page type.
- The schema.org validator for correct syntax.
- The “Enhancements” reports in Search Console after publishing, to catch errors across the site.
Frequently asked questions
Does structured data raise my ranking?
It doesn’t guarantee a higher ranking, but it helps Google understand the page, and may make it eligible for a richer display that raises click-through.
Does it help me appear in AI answers?
Google says no special markup is required for AI features, and that structured data must match the page. It’s part of a clear, understandable site; good content remains the foundation.
How many types should one page have?
Whatever genuinely describes the page: an article page, for example, carries Organization, Article and BreadcrumbList, plus FAQPage if it has questions. Don’t add types that don’t match the content.
The bottom line
Structured data translates your page for machines, provided it says the same as the page. Choose the right types, generate them from the content itself, and test them. If you want a website built this way, have a look at our website service.
