Schema Markup Implementation: A Practical JSON-LD Guide

Schema markup is structured data that describes the content of a page using a shared vocabulary. For example, an article can identify its headline, author, and publication date. Google can use supported structured data to understand a page and make it eligible for certain enhanced search appearances (S1).

Valid markup does not guarantee a rich result or a higher ranking. Start with accurate information that people can already see on the page, then add only the properties that describe it. Do not add reviews, prices, or authors that you cannot substantiate.

Choose a type that describes the page#

Browse Google's structured data gallery to see which search features it supports, then read the specific documentation for your content type. Schema.org includes many types that Google does not use for a rich result, so valid Schema.org syntax alone is not proof of eligibility.

A page may contain more than one relevant type, but adding every possible type will not make it more eligible. Some search appearances have additional restrictions; consult the current feature documentation before choosing a type.

Create a JSON-LD example#

JSON-LD is a way to express linked data in a script element in the HTML document. The following is a fictional example, not a claim about a real article or person. Replace every value with information from your own page.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "How to maintain a garden",
  "author": {
    "@type": "Person",
    "name": "Jordan Lee"
  },
  "datePublished": "2026-01-15",
  "dateModified": "2026-10-28"
}
</script>

Keep the date modified aligned with the actual substantive revision shown to readers. If an article has no verified author, do not invent one. If your platform generates markup automatically, inspect its output before adding a second script to avoid conflicting descriptions.

The JSON-LD generator can help prepare a starting point. Check that every value is correct, then paste the output into your page's HTML or configure your publishing platform to render it. Tool output is a draft, not a substitute for testing your published page.

Validate before and after publishing#

1. Check the feature requirements. Review the relevant Google documentation for required and recommended properties and any content-specific rules.
2. Test the rendered page. Use the Rich Results Test to check Google-supported markup. Use the Schema Markup Validator for broader Schema.org syntax. A pass in one test does not guarantee a rich result.
3. Inspect access and indexing. Confirm that the public page returns a successful response, is crawlable, and is not marked noindex. Search Console's URL Inspection tool distinguishes the indexed version from a live test (S3).
4. Monitor changes. Where Search Console provides an enhancement report for the type, review errors and affected URLs after release. Recheck markup when the underlying content or the platform's requirements change.

Common mistakes#

• Markup contradicts the page. Keep facts, prices, availability, dates, and reviews synchronized with the visible content.
• A template emits irrelevant markup. A business listing is not a product review, and a normal article is not automatically an FAQ page.
• A valid script describes inaccessible content. Crawlers still need access to the page. A test cannot guarantee indexing or display.
• Generated data is treated as verified. Review suggested values manually; never fabricate authorship, ratings, business details, or claims.
• AI search promises are attached to schema. Google says no special schema is required for AI Overviews or AI Mode (S4).

Sources#

• S1: Google Search Central: Understand how structured data works
• S2: Google Search Central: Structured data general guidelines
• S3: Search Console Help: URL Inspection tool
• S4: Google Search Central: AI features and your website

Last updated: October 28, 2026.