Schema Markup Validator: How to Test Structured Data (and Why It Affects AI Overview Citations)

A schema markup validator checks whether the structured data on your page — the JSON-LD or microdata that tells Google "this is an Article," "this is a Product," "this is a Recipe" — is written correctly and matches what's actually on the page. If it's broken, Google can still crawl and index your content fine, but it loses the extra context that makes rich results, and increasingly AI Overview citations, possible.

Most bloggers only think about structured data once, right after they add an SEO plugin, and never check it again. That's a problem, because structured data breaks quietly. A theme update, a plugin conflict, or a manual edit to a template can strip out a required field, and nothing in your normal Google Search Console workflow will flag it as urgently as a dropped page.

What a Schema Markup Validator Actually Checks

A validator parses your page's structured data against the schema.org vocabulary and, in Google's case, against Google's own eligibility requirements for rich results. It checks three separate things:

• Syntax — is the JSON-LD valid JSON at all? A missing comma or unclosed bracket can invalidate the entire block.
• Required fields — does an Article have a headline, image, and datePublished? Does a Product have a price and availability? Google publishes the required and recommended fields for each type, and missing a required one disqualifies you from that rich result.
• Content match — does the structured data actually reflect what's visible on the page? Google has been explicit that structured data describing content users can't see is treated as spam, not a technical error.

That third point trips up more sites than the first two combined. It's common to see an old FAQ schema block left in the code after the visible FAQ section was removed from the page — technically valid JSON, completely mismatched with reality.

The Free Tools Worth Using

You don't need a paid tool for this. Three free options cover almost everything:

| Tool | Best for | Limitation | |---|---|---| | Google Rich Results Test | Checking eligibility for specific rich result types (FAQ, Article, Product, Recipe) | Only tests types Google currently supports as rich results | | Schema Markup Validator (schema.org community tool) | Full schema.org vocabulary validation, including types Google doesn't turn into rich results | Doesn't tell you whether Google will actually use the markup | | Search Console Enhancements report | Site-wide monitoring of valid vs. invalid items over time | Only shows types Google has decided to track; lags behind a fresh page by a few days |

The workflow that catches the most real problems is to use the first two on individual URLs when you publish or edit a page, then rely on the Search Console Enhancements report for ongoing, site-wide monitoring — that's the one that will actually tell you if something broke last week on fifty pages at once, not just the one you happen to be looking at.

Common Structured Data Errors (And What They Actually Mean)

A few errors show up constantly enough that it's worth knowing them ahead of time:

"Missing field 'image'" on Article or Product schema almost always means the featured image field in your CMS isn't mapped to the schema output, not that the image is actually missing from the page.

"Invalid value for field 'datePublished'" usually means the date format isn't ISO 8601. A theme showing "September 9, 2026" to readers still needs to output 2026-09-09 in the structured data.

"Duplicate field" happens when a plugin and a manually-added script both inject the same schema type on one page. Google doesn't merge them — it just gets confused about which one to trust, and confused signals get ignored rather than penalized.

None of these are indexing-breaking on their own. What they cost you is eligibility for the presentation upgrades — stars, FAQ dropdowns, product prices — that structured data exists to unlock in the first place.

Why Structured Data Affects Whether AI Overview Cites You

This is the part that's changed the calculus on fixing these errors. Google's AI Overview and AI Mode pull from pages it can parse cleanly and attribute confidently, and structured data is one of the clearest confidence signals available: it's an explicit, machine-readable statement of what a page is and what claims it's making, rather than something Google has to infer from prose.

That doesn't mean adding schema guarantees a citation — plenty of uncited pages have perfect markup, and plenty of cited ones have none. But when a page has a factual error in its structured data, or contradicts its own schema, it's a straightforward reason for Google to trust a different source instead. Clean structured data doesn't earn you a citation. Broken structured data can quietly cost you one you'd otherwise have gotten.

If you want the fuller picture on what actually drives AI Overview citations beyond just structured data, How to Get Cited in Google AI Overview covers the rest of the checklist, and the AI Overview SEO Audit Checklist walks through the technical side in more depth.

A Simple Testing Workflow You Can Repeat Every Month

You don't need this to be a heavyweight process. A workable monthly routine looks like:

1. Run your 3-5 highest-traffic pages through the Rich Results Test. Fix anything flagged as an error (not just a warning).
2. Check the Search Console Enhancements report for any type that shows a sudden jump in "invalid" items — that's usually a template or plugin change, not content drift.
3. Spot-check one recently published page with the full schema.org validator, since Search Console only tracks types it's decided to surface as rich results.

This is the same rhythm Traffic Monitor's Post SEO Audit uses for on-page checks generally — catching drift early, on a schedule, instead of only noticing when a rich result disappears from search results months later. If you're already tracking whether your pages get cited in AI Overview, pairing that with a quick structured data pass each month closes one of the more avoidable gaps between "eligible" and "actually cited." You can log in and run a Post SEO Audit alongside your next structured data check if you want both in one pass.

Structured data errors are rarely dramatic. They're the kind of thing that sits broken for months because nothing in your traffic dashboard screams about it — until you're wondering why a competitor with a thinner article is the one showing up with stars, or the one AI Overview decided to quote.