Blog
What Is Schema Markup?
What is schema markup?
Schema markup is a bit of code you add to a page that spells out what your content actually means, which text is a price, which is a review score, which is the author, which is an address. Plain HTML only tells a browser how to display text. Schema tells search engines what the text is, so they can read your page the way a person does instead of guessing.
Think about a string like “4.8.” To you it’s obviously a star rating. To a search crawler it’s just two digits and a dot until something tells it otherwise. That “something” is schema markup: a shared vocabulary from Schema.org that labels each piece of your content so a machine reads it correctly the first time, no guessing.
That matters more now than it did five years ago. It’s the difference between a plain blue link and a result with stars, prices, and images baked in. And it’s increasingly how ChatGPT, Perplexity, and Google’s AI Overviews decide whether to pull an answer from your site or someone else’s.
How does schema markup actually work?
When a search engine crawls your page, it reads the schema code, matches the labelled entities (your business, a product, an article, an author) against its own knowledge graph, and uses that to decide how to display you and how much to trust you.
The mechanics are simple. You drop a small script onto the page. Inside it, you name the thing the page is about and describe its properties: an Article has an author, a datePublished, a headline; a Product has a price, a currency, an aggregateRating. Google’s crawler finds that script, checks it’s valid, and maps those properties to its own index.
The payoff is two-fold. First, accuracy: you’ve removed the guesswork, so Google indexes your page as exactly what it is. Second, eligibility. Valid schema is what makes a page qualify for rich results, the review stars, the FAQ dropdowns, the recipe cards, the breadcrumb trails. No schema, no rich result. It’s a gate.
This is also the connective tissue behind entity-based SEO. Structured data is how you tell Google, in plain machine terms, that your brand is a specific entity with specific expertise, rather than leaving it to infer that from scattered HTML.
Which format should you use? JSON-LD vs microdata vs RDFa
There are three ways to write structured data. In practice, one of them wins for almost everyone.
| Format | How it’s added | Google’s stance | Hassle |
|---|---|---|---|
| JSON-LD | A self-contained <script> block, usually in the page head | Recommended | Low. Sits apart from your visible HTML |
| Microdata | Attributes woven into your existing HTML tags | Supported | High. Clutters your markup |
| RDFa | XML-style attributes inside HTML elements | Supported, older | High. Breaks easily on redesigns |
Use JSON-LD. Google recommends it, and the reason is practical: because it lives in its own script block instead of being tangled into your visible markup, you can update a price or an author or an address without touching your layout and without risking a broken page. Microdata and RDFa both smear the data across your HTML, so every redesign becomes a chance to break something.
How do you test schema markup?
You test schema markup by running your live URL or your raw code through a validator that checks it against Schema.org rules and Google’s rich-result requirements. It flags missing properties, typos, and wrong data types before they cost you.
Two tools do the job. Google’s Rich Results Test tells you whether a page qualifies for a specific rich result. The Schema.org validator checks that your syntax is well-formed in general. Run one before anything goes live.
Why bother? Because one missing field silently disqualifies you. Leave the currency out of a product’s price, or the image off an article, and Google won’t reject the page, it’ll just quietly decline to show the rich result, and you’ll never see the stars you were expecting. Validation is how you catch that before publish, not three weeks after.
To run Google’s Rich Results Test:
- Open the tool on Google’s search developer site.
- Paste your live URL, or switch to the Code tab and paste the raw JSON-LD.
- Run the test.
- Read the report: it lists detected items and says whether each is valid and eligible.
- Fix any warnings about missing properties and re-test.
Keeping your structured data clean is part of the same technical hygiene as fast page loads: unglamorous, easy to skip, and quietly decisive for how you rank.
How do you fix a schema validation error?
You fix a schema error by finding the exact property the validator flagged, checking your JSON-LD against the official Schema.org definition, and correcting whatever’s wrong: a missing required field, a mismatched data type, or a broken bracket.
Most errors are boring and mechanical. A quotation mark that wasn’t escaped. A missing closing bracket on a nested object. A rating passed as text (“four”) when the schema wants a number (4). The fix is almost always one of these:
- Find the line. The validator gives you a line number or property name. Start there.
- Match the type. Strings, numbers, and arrays have to match what the schema type expects. A price is a number, not a word.
- Check the nesting. A
Publisherinside anOrganizationhas to sit in the right place in the hierarchy, or the whole block reads as invalid. - Clear cache and re-test. Redeploy, clear any server cache, run the URL through again to confirm it’s clean.
Get it right and the reward is real: accurate indexing, eligibility for rich results, and a better shot at being the source an AI engine cites.
Does schema markup improve your rankings?
Not directly. Schema markup is not a ranking signal on its own, so adding it won’t push you up the results page by itself. What it does is help Google understand and trust your content, and unlock rich results that pull far more clicks, which is often worth more than a position or two anyway.
This trips people up, so it’s worth being blunt about. You can’t schema your way to the top. But a result with review stars and a price sitting at position three will frequently out-earn a plain link at position one, because it takes up more space and answers more of the question before the click. Schema wins you attention and trust, not rank. Those tend to translate into rank over time regardless.
Frequently Asked Questions
What is schema markup in plain terms?
It’s a small piece of code you add to a web page that translates your content into something search engines read directly, so they know which text is a product, a price, a review, an author, or an address instead of having to guess from the layout.
Why is JSON-LD the preferred format?
Because it sits in its own script block, separate from your visible HTML. That makes it easy to add, update, and scale without breaking your page layout, which is exactly why Google recommends it over microdata and RDFa.
Does schema markup guarantee better rankings?
No. It isn’t a direct ranking factor. It helps search engines index your content accurately and makes you eligible for rich results, and those rich results tend to lift click-through rates, which is where the real gain comes from.
Which businesses benefit most from schema markup?
E-commerce stores (product and review schema), local service businesses (location and hours), publishers and blogs (article and author schema), event organizers, and any B2B company that wants its expertise recognized as a distinct entity all get outsized value from it.
How often should you check your schema?
Any time you change your site’s templates, CMS, or code, since a redesign can silently break structured data, plus a routine check every quarter as part of general SEO housekeeping. Broken or missing schema is one of the standard findings in a technical audit, alongside crawl errors, indexing gaps and page speed.
Sources: