Testing and Validating Your Schema
Schema you haven't tested is schema you're hoping works. Here's how to check it in a few minutes, find the usual problems, and pull everything from this course into one connected graph.
Two tools, two different questions
There are two free testing tools, and they answer different questions. You'll want both.
Rich Results Test
search.google.com/test/rich-results
Asks: is this page eligible for Google's rich results? It only reports on the types Google uses for rich results, and checks them against Google's own required and recommended properties.
Schema Markup Validator
Asks: is this valid Schema.org markup? It reads every type, including ones Google doesn't show as rich results, like WebPage and Service, and shows the full structure.
Both accept a live URL or a pasted code snippet. Our routine: paste code into the Schema Markup Validator while you're writing, to catch typos and see that your @id links hold together. Then, once it's live, run the URL through the Rich Results Test to see what Google makes of it.
Don't panic if the Rich Results Test says it found no items on a page that only has WebPage or Service markup. That means "nothing here is a rich result type," not "your markup is broken." Check it in the validator instead.
Search Console: testing over time
The testing tools check one page at a time. Google Search Console watches your whole site. Once Google has crawled pages with rich-result markup, look in the left-hand menu for the section with reports for those types (it has been labeled Enhancements, and some types appear under other headings). Each report lists valid items, items with issues, and the pages affected.
- Errors make that item ineligible for the rich result. Fix these first.
- Warnings are usually missing recommended properties. The item can still be eligible; adding them gives Google more to work with.
- After fixing a problem, use the report's validate fix option so Google rechecks the affected pages.
For a single page, the URL Inspection tool in Search Console also shows which structured data Google detected on its last crawl, and can test the live version.
Find every block on the page
Before you trust a test result, see what's actually on the page. Open it in your browser and View Page Source (right-click, or Ctrl+U on Windows, Cmd+Option+U on a Mac). Search for ld+json and count the blocks. Two Organization blocks is a red flag.
View Source shows the HTML the server sent. If a block is added by JavaScript, it won't be there. To see what the browser ended up with, open DevTools and paste this into the Console. It lists every JSON-LD block on the page, and flags any that aren't valid JSON:
document.querySelectorAll('script[type="application/ld+json"]').forEach((s, i) => {
try { console.log(i + 1, JSON.parse(s.textContent)); }
catch (e) { console.warn(i + 1, "Invalid JSON:", e.message); }
});
Common errors, and how to fix them
| Problem | What it looks like | Fix |
|---|---|---|
| Invalid JSON | The tools find nothing, or report a parsing error | Look for trailing commas, missing commas, curly quotes, or comments (Lesson 2). One error hides the whole block. |
| Missing required property | An error in the Rich Results Test, like a missing startDate on an Event | Add it, if it's true and visible on the page. Each lesson lists the required properties. |
| Relative URLs | "url": "/services/" | Use full URLs: "https://fernandink.example/services/". |
| Doesn't match the page | Schema shows a price, rating, or date the visitor can't see | Change the markup or the page so they agree. Markup must describe visible content. |
| Duplicate Organizations | A theme says "Fern and Ink Studio," a plugin says "Fern & Ink" | Pick one source, use one name and one @id everywhere, and switch the other off. |
| Wrong date format | "November 16, 2026" or "11/16/2026" | Use ISO 8601: "2026-11-16", or with a time and time zone, "2026-12-05T18:00:00-06:00". |
Broken @id links | A reference like { "@id": "...#organisation" } that matches nothing | Copy and paste @id values instead of retyping them. They must match exactly. |
"Valid" isn't the finish linePassing both tests means your markup is eligible, not that Google will show a rich result. And a test can't tell whether the markup is true. Only you can check that it matches the page.
Putting it all together
Here's what everything in this course adds up to: one connected @graph for Fern & Ink's letterpress printing page. The Organization and WebSite are the same on every page. The WebPage and BreadcrumbList change per page. Every piece links to the others by @id.
{
"@context": "https://schema.org",
"@graph": [
{
"@type": "Organization",
"@id": "https://fernandink.example/#organization",
"name": "Fern & Ink",
"url": "https://fernandink.example/",
"logo": "https://fernandink.example/images/logo.png",
"telephone": "+1-612-555-0142"
},
{
"@type": "WebSite",
"@id": "https://fernandink.example/#website",
"name": "Fern & Ink",
"url": "https://fernandink.example/",
"publisher": { "@id": "https://fernandink.example/#organization" }
},
{
"@type": "WebPage",
"@id": "https://fernandink.example/services/letterpress-printing/#webpage",
"url": "https://fernandink.example/services/letterpress-printing/",
"name": "Letterpress Printing | Fern & Ink",
"isPartOf": { "@id": "https://fernandink.example/#website" },
"breadcrumb": { "@id": "https://fernandink.example/services/letterpress-printing/#breadcrumb" },
"dateModified": "2026-11-16"
},
{
"@type": "BreadcrumbList",
"@id": "https://fernandink.example/services/letterpress-printing/#breadcrumb",
"itemListElement": [
{ "@type": "ListItem", "position": 1, "name": "Home", "item": "https://fernandink.example/" },
{ "@type": "ListItem", "position": 2, "name": "Services", "item": "https://fernandink.example/services/" },
{ "@type": "ListItem", "position": 3, "name": "Letterpress Printing" }
]
}
]
}
Paste this into the Schema Markup Validator and you'll see four items, each pointing to the others. Paste it into the Rich Results Test and you'll likely see only the breadcrumbs reported, because that's the one rich result type here. Both results are correct. From here, you'd add whatever the page is actually about, like a Service (Lesson 8), into the same @graph.
The five-minute schema review
Run this on any page, yours or a client's. Five minutes, no special tools.
- Find it. View source, search
ld+json, and count the blocks. Note what's generating each one. - No duplicates. One Organization, one WebSite, with the same name and
@ideverywhere. - Valid JSON. Paste each block into the Schema Markup Validator. No parsing errors.
- Right type. The main
@typematches what the page is really about. - Matches the page. Names, prices, dates, and ratings in the markup are visible on the page.
- Full URLs and proper dates. No relative links; dates in ISO 8601.
- Connected.
@idreferences point to items that exist, spelled exactly the same. - Google's view. Run the live URL through the Rich Results Test and fix any errors.
- Over time. Check Search Console's reports after the next crawl.
Try it
Run the five-minute review on your own homepage. Write down every problem you find, then fix the worst one: usually a duplicate Organization or invalid JSON. Then run it on one inner page, like a blog post or a service page. Inner pages are where markup tends to get forgotten.
Check yourself
- I know the Rich Results Test checks Google eligibility and the Schema Markup Validator checks Schema.org validity.
- I know where Search Console reports structured data issues across my site.
- I can find every JSON-LD block on a page, including ones added by JavaScript.
- I can spot and fix the common errors: bad JSON, missing properties, relative URLs, mismatches, duplicates, and date formats.
- I can build one connected
@graphfor a page.
That's the course. You can read JSON-LD, write it, put it on a site, and prove it works. Go mark something up.
Schema is one item on a longer list. See where it fits in a full site review: Technical SEO Audit Checklist: 50 Checks in Priority Order.