The single most useful SEO structured data testing tool is Google’s Rich Results Test, because it answers the only question that pays: will this markup actually produce a rich result in Google Search. Everything else validates syntax. Syntax is table stakes.
The old Structured Data Testing Tool that most guides still reference was retired by Google in 2021 and handed to schema.org, where it lives as the Schema Markup Validator. That split confuses people constantly, so here is the clean breakdown of what each tool does, when to reach for which, and what the results actually mean.
The three tools, and what each one is for
Three tools cover essentially every structured data need, and they answer different questions.
- Rich Results Test (search.google.com/test/rich-results) — tests a live URL or a code snippet and tells you which Google rich result types the page is eligible for. This is the one that matters for search visibility.
- Schema Markup Validator (validator.schema.org) — checks your markup against the full schema.org vocabulary. It flags invalid properties and types that Google’s tool ignores because Google does not use them. Useful for non-Google consumers of your data and for general correctness.
- Search Console enhancement reports — the only source that shows how your markup behaves across your whole site at scale, on pages Google has actually crawled, over time.
The practical workflow: build with the Schema Markup Validator, confirm eligibility with the Rich Results Test, monitor at scale in Search Console. Skipping the third step is the common mistake. A page can pass both testers today and start throwing errors next month when a plugin update changes the output.
Errors, warnings, and what actually blocks a rich result
Google’s tool separates issues into two tiers, and the distinction is worth internalizing because teams routinely waste days on the wrong one.
An error means a required property is missing or malformed, and the page is not eligible for that rich result. A Product without an offers, review, or aggregateRating property will not show pricing in the SERP. Fix every error.
A warning means a recommended property is absent. The rich result can still appear, but it may render with fewer features. A Recipe missing prepTime still qualifies; it just shows less detail in the carousel. Fix warnings on your highest-traffic templates and ignore them elsewhere.
One more thing the testers cannot tell you: eligibility is not a guarantee. Google decides per query whether to render a rich result at all. Valid markup buys you a ticket, not a seat.
JSON-LD, Microdata, and RDFa: pick one
Google supports all three formats and explicitly recommends JSON-LD. Use JSON-LD. It sits in a single <script type="application/ld+json"> block, usually in the head, completely separate from your visible HTML. That separation is the entire point — you can change your page layout without breaking your markup, and you can inject it from a tag manager or a CMS field.
Microdata and RDFa interleave attributes into your HTML elements. They work, but every template edit risks orphaning a property, and debugging means reading markup scattered across a page. If you inherited a Microdata implementation and it validates cleanly, leave it. If you are starting fresh, JSON-LD is not a close call.
One caveat worth knowing: if your JSON-LD is injected by client-side JavaScript, Google will usually still pick it up after rendering, but it adds latency and a failure mode. Server-render it where you can.
The schema types worth your time
Schema.org defines over 800 types. Perhaps a dozen produce visible search features, and your site probably needs three or four of them.
- Article / BlogPosting — supports Top Stories eligibility and article-style presentation. Include
headline,datePublished,dateModified,authorwith a realPersonorOrganization, andimage. - Product with
offers— price, currency, and availability in the result. The highest-value markup in ecommerce, full stop. - FAQPage — Google narrowed FAQ rich results in 2023 to well-known authoritative government and health sites. Mark up your FAQs anyway for machine readability, but do not budget for a CTR lift.
- BreadcrumbList — replaces the raw URL in the result with a readable path. Cheap to implement, easy win.
- LocalBusiness and Organization — feeds knowledge panel and entity understanding. Include
sameAslinks to your verified social and directory profiles.
How to test properly before you ship
Test the rendered page, not your source template. Paste the live URL into the Rich Results Test rather than the code snippet whenever the page is publicly reachable, because the URL test runs Google’s actual renderer and catches JavaScript problems that snippet testing hides.
For pages behind authentication or on staging, use the code snippet mode, then retest against the live URL after deployment. A markup block that validates in isolation and breaks in production is nearly always a templating issue: an unescaped quote in a product name, a null value where a date should be, or a description field containing raw HTML.
Test one representative URL per template, not one per page. If your product template is correct, all 8,000 products are correct until someone changes the template. Then spot-check a handful of edge cases: the product with no reviews, the article with two authors, the out-of-stock item.
Monitoring after launch, where most implementations rot
Structured data degrades silently. Nobody notices when a CMS update strips the author field, because the page looks identical. You find out six weeks later when your article results stop showing bylines.
Search Console’s enhancement reports are the safety net. Each supported type gets its own report with valid, warning, and error counts plotted over time. Check them monthly and treat any sudden step change as a deployment incident — those cliff edges almost always trace back to a specific release.
Pair that with your regular crawl. A full-site crawl that extracts and flags structured data blocks catches drift on pages Google has not recrawled yet, which is the difference between fixing a bug in two days and finding it in two months. SEO Rocket’s deep crawls surface concrete per-page evidence — actual titles, URLs, and H1 text alongside technical issues — so a broken template shows up as a specific list of affected URLs rather than a count.
Common mistakes worth avoiding
Do not mark up content that is not visible on the page. Google’s guidelines are explicit, and invisible-content violations trigger manual actions that remove all rich results for your site. If the review does not appear to a human visitor, it does not belong in your markup.
Do not invent aggregate ratings. Self-serving reviews on your own products, ratings pulled from nowhere, and star markup on pages with no actual reviews are the fastest route to a structured data penalty. Google has enforced this aggressively.
Finally, do not treat schema as a ranking lever. It is a presentation lever. A page ranking at position 18 with immaculate Product markup is still a page at position 18. Fix relevance and authority first, then use structured data to win more clicks from the position you have earned.
