Free Schema Markup Validator (JSON-LD Checker)

What is a schema markup validator?

A schema markup validator reads your JSON-LD structured data and reports missing required fields, wrong values and structural mistakes, so you can fix them before search engines read the page. This one also checks the links between items in an @graph and, for a live URL, whether the marked-up text appears on the page.

What this validator tests

Schema checker: supported types and what each one needs

This schema checker has built-in checklists for the types below. For each one it flags missing required fields as errors and missing recommended fields as warnings. Other types are still parsed and shown, but without a checklist.

These lists are this tool's checklists, based on the fields that matter for common rich results. Google's requirements vary by feature and change, so confirm them in Google's structured data gallery before you rely on a result.

JSON-LD validator: the value rules it checks

A field can be present and still be wrong. This validator checks the format of values, not just whether a field exists:

It also warns about headlines over 110 characters, publish or modified dates in the future, and phone numbers that look too short.

Common schema markup errors and how to fix them

1. Trailing comma or curly quotes

A comma after the last item, or quotes copied from a word processor, makes the whole block invalid JSON. The tool repairs these to be able to check the rest, then tells you to fix the original.

Wrong:  { "@type": "Product", "name": "Widget", }
Right:  { "@type": "Product", "name": "Widget" }

2. Price outside the offers object

The validator checks exact property paths. A price placed directly on a Product does not count as an offer, so the report says the Product needs at least one of offers, review or aggregateRating.

Wrong:  { "@type": "Product", "name": "Widget", "price": 19.99 }
Right:  { "@type": "Product", "name": "Widget",
          "offers": { "@type": "Offer", "price": 19.99, "priceCurrency": "USD" } }

3. A price with a currency symbol

Wrong:  "price": "$19.99"
Right:  "price": 19.99, "priceCurrency": "USD"

4. An http context or a missing @context

Wrong:  "@context": "http://schema.org"
Right:  "@context": "https://schema.org"

5. A date in the wrong format

Wrong:  "datePublished": "2026/10/01"
Right:  "datePublished": "2026-10-01"

6. A rating outside its scale, or a review count of 0

Wrong:  "aggregateRating": { "ratingValue": 9, "reviewCount": 0 }
Right:  "aggregateRating": { "ratingValue": 4.6, "reviewCount": 128 }

Use real ratings from reviews visitors can see. If there are no reviews, remove the whole rating block.

7. A nested item with no @type

Nested items such as an offer, address, rating or step should say what they are. The tool names the type it expects, and its corrected code adds it for you.

To use the auto-fix, paste your code, read Corrected code, then copy it and fill in any real data it could not invent. Only safe fixes are applied automatically: straight quotes, comments, trailing commas, single quotes, https context, date and price format, uppercase currency and full schema.org URLs for enum values.

How the schema health score is calculated

The score starts at 100. Each error takes off 12 points and each warning takes off 3, and the score never goes below 0. So one error and two warnings gives 100 - 12 - 6 = 82. The number is a quick summary of the report, not a Google score, and it says nothing about whether a page will get a rich result.

@graph and @id: how connected schema is checked

Many sites and SEO plugins output one @graph that holds several items, such as a WebSite, an Organization and an Article, and let them refer to each other by @id. The validator checks each item in the graph on its own, then checks the links between them:

• Each @id should be defined once. A duplicated @id is an error.
• A reference such as { "@id": "https://example.com/#organization" } should point to an item that is defined. When it does not, you get a warning on pasted code and an error on a live page.
• Repeated Organization, WebSite or FAQPage items get a warning, because a plugin and a hand-written block often both output the same one.

When you paste only part of a page, a reference can look undefined because its target sits elsewhere on the page. That is why this check is a warning for pasted code.

How to validate JSON-LD

1. Paste your code, or switch to Check a live URL and enter a page address.
2. Press Validate. Start with the Fix first list.
3. Open each item to see what passes and what is missing.
4. Copy the corrected code if fixes were applied, then add the real data for missing fields.
5. Confirm the live page in the Google Rich Results Test.

Valid schema but no rich result: why

Valid markup makes a page eligible. It does not make Google show a rich result. Google decides based on the page, the query and its own policies. Two types in this tool carry a note for that reason:

• FAQPage. Google stopped showing FAQ rich results in search on May 7, 2026. The markup is still valid, and Google says it can stay in place.
• HowTo. Google cut back HowTo rich results in 2023, first to desktop only, so do not plan around them. The markup is still valid for other readers.

Other common reasons a valid page shows no rich result: the marked-up text is not visible on the page, the page has a quality or policy problem, the markup describes the wrong thing, or Google has not recrawled the page yet. In URL mode this tool warns when marked-up text, such as a headline or product name, is not found in the page's visible text.

Google's guide to generative AI features also says structured data isn't required for generative AI search and that there is no special schema.org markup to add. Valid markup is still worth having for the rich results Google supports and for clarity.

This validator vs the schema.org validator and Google's Rich Results Test

These tools answer different questions, so use more than one.

A good workflow is to paste your code here, fix what it finds, confirm the vocabulary in the schema.org validator, and then test the live page in the Rich Results Test. To build the code first, use the Schema Markup Generator.

What this schema validator does not do

• It reads JSON-LD and microdata. It does not check RDFa and tells you when it finds it.
• It does not confirm eligibility for a Google rich result. Use the Rich Results Test for that.
• In URL mode it reads the HTML your server sends. Schema added by JavaScript after the page loads is not seen, and the report says so when it finds none.
• Only the types in the table above have checklists. Other types are parsed and shown without checks.
• It checks one page or one pasted block at a time.
• Pasted code is checked in your browser and is not sent anywhere. In URL mode our server fetches the public page for you and does not keep it.
• It cannot tell whether your markup is true. Every value must match what visitors can see.

Schema validator FAQ

Built by Vansh Bohra, an SEO specialist in Ahmedabad, India. Last updated October 2026.