How to Build and Check JSON-LD Schema Markup

Schema markup describes the entities and facts already present on a page. Schema.org supplies the vocabulary; search engines decide which combinations support their own features. Google recommends JSON-LD for maintainability, but valid JSON-LD alone does not make a page eligible for a rich result. Required properties for the chosen Google feature, crawlability, policy compliance, and correspondence with visible content all matter, and even a fully correct implementation carries no display guarantee.

The practical review has three gates. A parser or Schema Markup Validator checks vocabulary and structure. Google's Rich Results Test checks whether a supported Google feature recognizes the markup and reports required-property problems. A human compares every claim with the rendered page and business records. Passing the first gate while failing the second is common for valid types without a Google enhancement; passing both while failing the third can create misleading markup and policy risk.

Choose the type that represents the page's main visible subject, not the search appearance you hope to obtain. Copy factual values from the page or source system, use absolute canonical URLs for identifiers and images, and add every required property before optional decoration. Generate the JSON-LD, inspect nested object types and date or duration formats, then run both the general Schema validator and the Google test relevant to that feature. After deployment, use URL Inspection and the applicable Search Console enhancement report.

Schema.org is the vocabulary reference, while Google Search Central is definitive for Google's supported features, required properties, quality rules, and testing workflow. Both sources are needed because vocabulary validity and search eligibility answer different questions.

Mark up an article whose visible page shows the headline `How QR Quiet Zones Work`, author `Maya Noor`, publication date `2026-08-10`, and a public 1200 × 630 hero image. Generate Article JSON-LD with the same values and canonical page URL. First remove the image to observe an eligibility warning, then restore it. Next change the JSON-LD author to a person absent from the page: the syntax may remain valid, but the truth gate fails. Finish by testing the deployed URL rather than only pasted code.

Rich-result readiness = valid markup ∩ supported feature requirements ∩ visible, accurate page facts ∩ crawlable, policy-compliant delivery. Removing any set from that intersection breaks readiness, and readiness still is not a promise of display.

A schema type can be valid at Schema.org while offering no Google rich result. FAQ rich results are generally limited to well-known authoritative government and health sites, and HowTo rich results were removed from Google Search, so neither should be sold as a universal visibility tactic. Product ratings and prices must describe the specific visible product and real offers. JavaScript-injected markup can be read, but rendering failures, blocked resources, canonical conflicts, or stale templates can keep deployed data from matching a local test.

Clear local errors, verify every claim against visible page content, then use the official Schema Markup Validator and Google's Rich Results Test where supported. Watch for one recurring error: publishing schema that claims content or offers not visible on the page.

Choose a supported type, enter only facts shown on the target page, generate or check the JSON-LD, and resolve local errors and warnings. Copy the script into the matching page, then validate the public URL with external tools after deployment.

Input: a schema type and its page-backed properties, or existing JSON-LD to inspect. Output: formatted markup, catalog-based diagnostics, and copy or download actions. The tool cannot verify ownership, truth, page visibility, crawling, indexing, or rich-result selection.

Generation, pasted-source parsing, validation, and file import run locally without fetching a page or uploading its data. The builder maps visible page facts into nested schema properties, while the checker parses existing JSON-LD and flags common required, format, and support issues.