Structured data: add accurate Schema.org markup

What is structured data?

Structured data expresses entities and relationships in a machine-readable format. Schema.org supplies a vocabulary; JSON-LD is one way to include it in HTML. Google supports a subset of types and properties for particular search features, not every Schema.org definition.

Accuracy comes before eligibility

Markup can help systems interpret content and make a page eligible for supported enhanced results. It is not a guarantee of a rich result or a general ranking boost. Not every page needs the same schema, so an inventory of detected types is an observation, not a task to add more.

Choose a type that fits visible content

Start with the page's main entity: an editorial article, a product offered for sale, or a breadcrumb trail. Check the current Google documentation for the feature you actually want before defining required fields. Do not invent review scores, authors, prices or availability to satisfy a validator.

1. Make the JSON-LD parse correctly

Inspect every application/ld+json block. Fix invalid JSON such as trailing commas, unescaped quotes or JavaScript comments. Use the Schema.org context and the correct property names. An @graph is a container for related nodes, not an excuse to add unrelated types.

2. Match facts and URLs to the page

Keep identifiers stable and URLs consistent with their intended entities. For a product offer, confirm that the shown price, currency and availability match the live offer. For an article, use the actual author and publication/modification dates; do not reset dates merely to appear fresh.

3. Describe a real navigation path

Breadcrumb items should follow the visible hierarchy, with consecutive positions and meaningful labels. Link each intermediate item to the actual category or section. Do not insert keyword-heavy levels that readers cannot navigate.

4. Only mark up questions readers can see

FAQ markup should describe genuine, visible questions and answers, including answers in expandable sections. Google stopped showing FAQ rich results in May 2026. Do not add FAQ schema expecting that retired search feature, or generate hidden questions solely for markup. Existing FAQ-content audit signals remain a reason to check factual consistency, not a promise of enhanced results.

5. Validate syntax, feature rules and the live page

Use Schema Markup Validator for vocabulary checks and Google's Rich Results Test for supported features. Then inspect the rendered page and review Search Console after deployment. A validator cannot establish that a claimed rating or offer is truthful; that remains your editorial responsibility.

Example: a real editorial article

This illustrates the structure, not a complete recipe for every Google feature. Substitute the real author, canonical URL and dates. Add other properties only when relevant and supported by the page.

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "How to adjust bicycle brakes",
  "author": { "@type": "Person", "name": "Example Author" },
  "datePublished": "2026-09-15",
  "dateModified": "2026-09-15",
  "mainEntityOfPage": "https://example.com/brake-adjustment"
}

Next.js: render data safely

Render JSON-LD in a server page using a native script element with type application/ld+json. Serialize data instead of building JSON with string concatenation. Escape < as its Unicode sequence before inserting the serialized value into HTML so text cannot close the script element.