Schema.org Structured Data: Which Schema, How to Test It
Which schema.org type suits which business, how to write it in JSON-LD, and how to check it is correct, built on Google's own rules.
rabbitclip teamPublished: 4 min read
Short answer
Structured data means marking up the information on a page, a product, a business, an article, using a standard vocabulary that search engines and AI tools can read directly. That vocabulary is schema.org, a shared resource maintained jointly by Google, Bing and the other major search engines.
The right markup can help a page appear with rich elements in search results: star ratings, prices, frequently asked questions. Wrong or incomplete markup is ignored at best, and can be treated as a spam signal at worst.
What does structured data actually do?
Structured data means adding a code block that repeats information already visible on the page, but in a form a machine can read; JSON-LD is the most common format and the one Google Search Central recommends, since it is easiest for site owners to implement and maintain at scale.
The code sits inside the page's HTML but stays invisible to the visitor; it lets a search engine say, effectively, 'this is a business page, here is its name, here is its address'.
Which schemas matter most for a business?
Three types tend to cover a business site's basics: Organization (brand identity: name, logo, social profiles), LocalBusiness (for premises with a physical address, opening hours included), and Article (for blog posts, with author and date).
A spa chain with its own LocalBusiness record on every branch page sends a far clearer signal than one central Organization record alone; each branch's address, phone number and hours can then be verified independently.
What are Google's general rules?
Google's general structured data guidelines say markup should not describe information invisible on the page; a person or product named in the schema needs to appear the same way in the page's own body text. Creating blank pages purely to carry markup is against the rules too.
Another rule is choosing the most specific type available, a Product, Service or FAQPage rather than a generic catch-all. Markup that breaks these rules can go undisplayed as a rich result even when the syntax itself is correct.
How is accuracy actually checked?
Google's Rich Results Test takes a page's URL or raw code and shows which schema is recognised and which fields are missing. It is worth running before the code goes live.
Once live, the Enhancements section in Search Console shows, across the whole site, which schema types are present and how many examples carry errors.
Common mistakes
A typical one we have seen in a workwear manufacturer's product catalogue is copying the same price or the same stock status across every item; once the underlying data moves on, the markup keeps showing something that is no longer true.
Another is leaving Organization details that contradict each other across pages, one phone number here, a different one there. Schema should repeat a single truth, not a different version on every page.
A useful decision rule follows from this: every field in the schema should have a matching statement in the page's own visible text. If it does not, either the page text needs updating or that field should not be in the schema at all.
How to verify it, step by step
A page's schema can go through a short checklist before it goes live.
- Add the JSON-LD block to the page's code and review the syntax by viewing source in a browser.
- Run the page's URL through the Rich Results Test and note which type is recognised and which fields are missing.
- Compare every field in the schema against the page's own visible text, one by one.
- Check the Enhancements section in Search Console a few days after the page goes live.
- Review other pages using the same schema type together, as a batch, rather than one at a time.
Structured data is the cheapest way to tell a search engine or an AI tool plainly what a page is about. Choosing the right schema, following Google's rules and testing regularly cover most of the work. Making this verification a pre-launch habit takes far less time than debugging the same schema after the fact. rabbitclip can scan a site's current schema and work out the gaps together.
FAQ
Does every page need Organization schema?
No. Google notes that Organization markup is enough in one place, such as the homepage, rather than on every page.
Should JSON-LD or microdata be used?
Google recommends JSON-LD for site owners because it can be added and maintained as one code block, separate from the page's visual markup.
Can wrong schema get a site penalised?
A direct penalty is rare, but non-compliant markup can simply be left out of rich results; serious misuse can be flagged as spam.
Does adding schema directly boost ranking?
No. Schema enriches how a page can appear; ranking is still decided by the content and overall quality of the site.
