Schema Markup Validator

Last updated: What’s new

Scan a live URL, or paste JSON-LD or HTML for a private browser-only check. Each JSON-LD and Microdata finding shows where it is and what to change. Use this diagnostic before Schema.org Validator and Google's Rich Results Test.

No signup required. Technical scan results may be used to produce aggregate website statistics. Research records do not include your name, email, IP address, or analytics ID. Privacy details

Check a live URL, or paste JSON-LD or HTML with Microdata:

Enter a domain or full URL. Domains use HTTPS by default. Scans structured data in the HTML returned by the server; it does not render JavaScript.

How the Schema Markup Validator Works

This schema and structured data checker reads JSON-LD and Microdata from pasted input or server-returned HTML, then reports each problem with its location and the change to make. Here's the process:

  1. Choose an input, fetch a public URL or paste JSON-LD or HTML for browser-only parsing without a server request.
  2. JSON-LD extraction, it finds all <script type="application/ld+json"> blocks, parses the JSON, and handles @graph arrays.
  3. Microdata parsing, it reads itemscope, itemtype, itemprop and itemref the way browsers do, including nested items, and takes each value from the right place: content on meta, href on links, src on images and datetime on time.
  4. Type detection, the tool lists the @type and itemtype values it finds, including types inside @graph and nested Microdata items.
  5. Common property checks, built-in rules flag selected gaps on frequently used types, for JSON-LD and Microdata alike. These checks are not a complete implementation of the Schema.org vocabulary or every Google search feature.
  6. Microdata structure and values, it flags itemprop outside any item, itemscope without itemtype, broken itemref ids, URLs taken from visible text, and dates that are not ISO 8601.

Why Schema Markup Matters for SEO and GEO

Structured data can make visible facts easier to interpret consistently. It is one layer of a broader search and entity strategy:

  • Search-feature eligibility, supported types with valid properties can make a page eligible for breadcrumbs, products, events, articles, and other enhanced results. Display is never guaranteed.
  • Entity clarity, stable identifiers and consistent Organization, Person, Product, or LocalBusiness facts can reduce ambiguity when they match the page and external profiles.
  • Maintenance, structured fields make stale prices, ratings, authors, and dates easier to audit systematically.
  • AI-search foundation, machine-readable facts may help systems interpret a source, but markup alone does not establish expertise or earn citations. Measure real prompts and cited URLs separately.

Schema works alongside your SEO and Open Graph tags ; together, they provide explicit metadata for different consumers. Need to generate a starting point? Use the Schema Markup Generator, then adapt it to the visible page and validate it. For a broader technical audit, the AI Search Readiness Checker reviews schema alongside crawler controls, headings, and standard discovery signals.

Want to go deeper? Our blog covers the full structured data topic: what schema markup is and why it matters for SEO explains the fundamentals and which types your site needs. Schema markup and AI citations discusses entity clarity and the limits of evidence linking markup to citations in AI search. And our guide to testing and validating schema markup walks through the complete validation workflow from pre-publish checks to quarterly audits.

Worked examples: a missing URL and broken Microdata

Use “Try a worked Organization example” above. The example has a context, type and business name, but no URL. The selected-property check flags that gap as a recommendation, not invalid JSON. Add the real canonical business URL when applicable, then validate again. The valid graph example demonstrates linked nodes and multiple types; the invalid example separates a malformed @type from a Google breadcrumb requirement.

“Try a Microdata example” loads a product snippet with four common template mistakes: the price sits outside its Offer, the product URL is read from link text instead of href, a date uses day/month/year, and itemref points to an id that does not exist. Each finding names the line and the exact change, such as moving itemprop="price" inside the Offer or using content="2026-12-31".

  1. Make the JSON parse. Fix syntax before interpreting missing properties.
  2. Match the page. Use the actual business identity and visible facts. Do not invent ratings, offers or qualifications to satisfy a checker.
  3. Validate the intended feature. Use Schema.org Validator for vocabulary, and Google’s Rich Results Test for supported Google features.
  4. Publish and rescan the URL. A successful pasted example does not establish what your live template emits.

For implementation and launch checks, follow our schema testing workflow. A passing test does not guarantee a search enhancement or AI citation.

Google Retired the Structured Data Testing Tool. Here's What to Use Now.

For years, the Structured Data Testing Tool was the way to check schema markup: paste a URL, see every type on the page, fix what's flagged. Google announced its deprecation in 2020, and in 2021 handed it over to Schema.org, where it survives as validator.schema.org without the Google-specific checks site owners relied on.

The replacements each cover only part of what the original did:

Validator What it checks The catch
Google Rich Results Test Schema types eligible for Google rich results Tests Google-supported search features; it is not a full Schema.org vocabulary validator.
validator.schema.org Syntax and vocabulary of all Schema.org types General vocabulary validation; use Google documentation for Google-specific search-feature requirements.
Search Console Rich result reports for your own verified site Only works on sites you own, and only after Google has crawled them.
This tool JSON-LD and Microdata from a live page or pasted code: parse and structure errors, Microdata value formats, and selected common-property checks Fast diagnostic only; not authoritative for the full Schema.org vocabulary or Google eligibility.

In short: use the Rich Results Test to confirm a specific rich result, and use a general schema checker like this one for a quick live-page inventory. No validator can predict an AI citation. For the complete validation workflow, read our guide to testing schema markup.

Common Schema Types, Chosen by Page

There is no universal checklist. Choose the narrowest types that accurately describe the visible page:

  • Organization, your business name, logo, URL, social profiles. The foundation of your online identity in search.
  • WebSite, site name, URL, and publisher identity.
  • BreadcrumbList, a real navigation path (Home → Category → Page) that can support breadcrumb display.
  • Article / BlogPosting, for editorial content with visible author and accurate publication dates.
  • FAQPage, only for visible FAQ content that complies with current search policies. Google retired FAQ rich results in May 2026; FAQPage remains a Schema.org type.
  • WebApplication, for tool pages (like this one). Describes the app, its category, pricing, and provider.
  • Product, for product pages. Google Product snippets require name and at least one of offers, review, or aggregateRating; merchant listings have additional requirements.
  • Service, for service pages. Describes what you offer, the provider, and the service area.

Schema Type Reference: Properties to Review

Different consumers and search features have different requirements. Use this as a starting reference, then confirm the current requirements for the specific feature in Google's documentation and validate the general vocabulary at Schema.org.

Schema type Use it on Common properties to review
Organization Site identity, commonly anchored from the home page @id, name, url, logo, and accurate sameAs profiles.
LocalBusiness Local business or store pages name, real address, telephone, openingHours, and geo when those facts are visible and accurate.
Article / BlogPosting Blog posts and news articles headline, image, datePublished, author. Include accurate dateModified and publisher where applicable. Google Article properties are recommendations, not a full-eligibility checklist.
Product + Offer Product and e-commerce pages name, image, plus offers with price and priceCurrency. Add aggregateRating or review for star ratings.
FAQPage Pages with a real FAQ section mainEntity as a list of Question items, each with a name and an acceptedAnswer (Answer with text).
BreadcrumbList Any page with a navigation path itemListElement as ordered ListItem entries, with position and name; include item (the URL) except that Google permits the final breadcrumb to omit it.
WebSite Home page (site-wide identity) @id, name, url, and a consistent publisher reference.

Article and breadcrumb rows rechecked on 2 October 2026; the wider reference was reviewed on 7 September 2026 against Google’s Article guidance, Product snippet requirements, breadcrumb requirements, and the FAQ retirement announcement. Article properties are recommendations; Product snippets accept offers, review, or aggregateRating; the final breadcrumb can omit item.

Requirements can change. Valid markup can make eligible content understandable; it does not guarantee a rich result, ranking, Knowledge Graph panel, voice answer, or AI citation.

What’s new

  • Interrupted compressed responses now finish with an explicit outcome instead of leaving the check waiting.

Next steps

Schema Markup Validator related tools and articles

Continue with the closest follow-up checks and guides based on this tool's topic, crawl intent, and optimization workflow.

How to test schema markup

What is schema markup?

Schema markup and AI citations

AI Citation Optimization: How to Get Your Content Cited by AI

Robots.txt for AI Search: Crawler Controls Without the Hype

GEO vs SEO: What Changes and What Stays the Same

Schema Markup Validator: FAQ

How do I check if a website has schema markup?
Enter the page URL above: the checker lists every JSON-LD block and Microdata item in the returned HTML, with its types. To check by hand, view the page source and search for application/ld+json or itemtype.
What do JSON-LD blocks, typed nodes, and Microdata items mean?
The report shows detected and successfully parsed JSON-LD blocks, and the typed JSON-LD nodes found directly, in a top-level array, or inside @graph. Microdata items are top-level itemscope elements, meaning items that are not a property of another item. The Microdata type inventory lists every itemtype in the HTML, including nested items such as an Offer inside a Product. These are inventory counts, not a quality score.
What do the issue messages validate?
The tool applies a small set of built-in checks to Organization, LocalBusiness, WebSite, article types, FAQPage, BreadcrumbList, software applications, and Product, whether the markup is JSON-LD or Microdata. Microdata also gets structure and value checks: itemprop outside any item, itemscope without itemtype, itemtype values that are not full URLs, itemref ids that do not exist, URL properties taken from visible text, and dates that are not ISO 8601. Syntax errors, structure errors, value errors, optional recommendations, and selected Google feature requirements are reported separately, with block numbers and property paths for JSON-LD and line numbers for Microdata. No issue means those checks passed, not that the full Schema.org vocabulary or a rich-result policy passed.
How should I fix a JSON-LD parse error?
The report identifies the script block number and the browser JSON parser message. Look for trailing commas, comments, unescaped quotes, invalid control characters, or markup mixed into the JSON. Fix the source block, publish it, and scan the live page again.
How are @graph, nested objects, and Microdata handled?
Top-level @graph nodes are listed separately, while nested authors, offers, and other objects remain properties of their owning item. Identical literal @id references can supply properties across blocks; external IDs and custom contexts are not fetched or expanded. Microdata is read with the HTML microdata rules: a nested itemscope becomes a property of the item around it, itemref adds properties from elements with the listed ids, and each value comes from the attribute browsers use, such as content on meta, href on links, src on images and datetime on time. Like Google, the checker also reads content on other elements. RDFa is not inspected.
Can I check FAQ schema here, and does it still show in Google?
Yes. FAQPage markup gets the built-in checks for mainEntity, each question name and its accepted answer text. Google stopped showing FAQ rich results on May 7, 2026, so valid FAQPage markup can describe visible FAQs but no longer produces an FAQ result in Google.
Does it check RDFa?
No. It reads JSON-LD and Microdata only. Use validator.schema.org to check RDFa.
Why can the result say no schema when I can see markup elsewhere?
The checker reads the fetched HTML response, up to 2 MB. Schema inserted only after client-side JavaScript runs may be absent from that response. Consent gates, authentication, bot protection, timeouts, or more than five redirects can also prevent the expected markup from being read.
Which validation tool should I use next?
Use this report to locate blocks and obvious gaps. Then use Schema.org Validator for general vocabulary conformance and Google's Rich Results Test for a Google-supported feature. Also compare every value with the visible page and check Search Console reports for your verified property.
What are the fetch and display limitations?
The request has a 12-second timeout, follows at most five redirects, and reads at most 2,000,000 bytes. Long property values are shortened for display. JSON-LD inspection is bounded to 100 blocks, 5,000 values and 64 nesting levels, with at most 200 typed nodes displayed. Microdata reading stops at 150,000 elements, 2,000 items or 32 levels of nested items, with at most 200 top-level items displayed. Limit notices mean the report is incomplete. The checker does not execute JavaScript or determine whether Google will show a rich result.
Are the checked URL and page markup stored?
The public URL is sent to Web Aloha's server to fetch and parse the HTML. When research collection is enabled, limited technical results, such as schema type and issue counts, may be used for research for up to 90 days to produce aggregate website statistics, then removed by daily cleanup. Research records use coded website and page keys, without raw URLs, HTML, schema property values, names, email addresses, visitor IP addresses, or analytics IDs. Pasted code is checked in your browser and never uploaded or included in research records. Ordinary hosting and analytics processing is described in our Privacy Policy.