Structured data and schema markup guide

Picture two search results side by side. One is plain blue text. The other shows a five-star rating, a price, and a photo. They point to similar pages, but one gets most of the clicks. The difference is structured data, and it is code you control.

Structured data is code, usually written in JSON-LD, that labels the content on your page so search engines understand what it means rather than just reading words. Schema markup uses the shared vocabulary from Schema.org. Add it correctly and you become eligible for rich results such as star ratings, FAQ dropdowns, and recipe cards.

What structured data does for you

Search engines read your page as text and images, and they infer meaning. Structured data removes the guesswork by stating facts directly: this number is a price, this text is an author, this block is a question and its answer. That clarity does two things. It makes your page eligible for rich results, the enhanced listings that stand out in search, and it helps engines index your content accurately. It will not raise your rankings on its own, but the extra space and detail in the result often lift your click-through rate.

Use JSON-LD

Schema markup comes in three formats: JSON-LD, Microdata, and RDFa. Use JSON-LD. Google recommends it, and it is the easiest to manage because it sits in a single script block in your page head, separate from your visible HTML. Microdata and RDFa weave attributes through your markup, which gets messy fast and breaks when you redesign. With JSON-LD you can add, edit, or remove markup without touching the content itself.

Schema types worth adding first

You do not need dozens of schema types; you need the few that match your content. Article markup suits blog posts. Product and Review markup drive the star ratings and prices that stand out on shopping results. Organization and LocalBusiness markup feed your knowledge panel and local results. FAQPage markup can add expandable questions under your listing, and it deserves its own care, so read our FAQ schema guide before you add it. Breadcrumb markup tidies up the URL line in your result.

How to add it

You have three routes depending on your setup. On WordPress, an SEO plugin can generate most schema automatically from fields you fill in. On other platforms, you can hand-write a JSON-LD block and drop it into the page template. If you would rather not touch code, Google Tag Manager can inject markup, though I prefer keeping it in the page source where it is easy to inspect. Whichever route you choose, the schema must match what a visitor actually sees; marking up content that is not on the page breaks the rules.

Test before you trust it

Never assume your markup works. Run every page through Google's Rich Results Test and the Schema.org validator. They parse your code, list the types they detect, and flag errors and warnings. Then watch the Enhancements reports in Google Search Console, which validate your live markup at scale and tell you when Google actually reads it. A single missing required field can disqualify a page from rich results without any obvious sign.

Keep your markup in sync with the page

Structured data rots quietly. When you change a price, rename a product, or edit an author, the visible page updates but hand-coded schema often does not, and stale markup that contradicts the page can cost you rich results. Tie your markup to the same data source that renders the page wherever you can, so one update changes both. If you manage markup by hand, add a recurring check to your calendar, because Google will keep validating it long after you forget it exists. On large sites, generate schema from templates rather than page by page, the same way you would manage any content at scale.

Structured data and AI answers

This part matters more each year. Clear, labelled content is easier for AI systems to parse, quote, and cite. Structured data helps engines pull the exact fact they need, which supports both featured snippets and the newer answer engines. If you want your pages surfaced in Google AI Overviews and cited by chatbots, schema is part of the groundwork, and our explainer on answer engine optimization shows where it fits into the bigger picture.

What schema will and will not do

Set your expectations honestly. Structured data makes you eligible for rich results; it does not force Google to show them, and Google decides case by case. It does not directly boost rankings, and stuffing fake reviews or marking up hidden content can trigger a penalty. Treated as accurate labelling rather than a ranking hack, it is one of the higher-leverage technical jobs on a technical SEO checklist. Our free SEO tools check your existing markup for errors, and the AEO and GEO checklist lays out the schema that helps AI engines cite you correctly.