Structured data coverage: pages that break your own norm

What a coverage outlier is

A page with no structured data is not always a defect. Plenty of pages legitimately carry none, and a rule that says so on every one of them is noise. A page that lacks the markup its own siblings carry is different: the site has already shown it can emit that type, and this page does not. That is usually a template or CMS gap, and it is always a specific thing to check.

This rule never asks whether a page has structured data in the abstract. It groups pages by their detected page type, works out what markup that group agrees on, and reports the pages inside the group that break the pattern.

What squirrel checks

The rule runs once per crawl and emits a single check named coverage-outlier. A page enters a comparison group only when its HTTP status is 2xx and it carries a page type that is neither empty nor unknown. The unknown bucket holds everything the classifier could not place, and its members are not siblings of each other.

• Skipped with the message Only N page(s) crawled; 10 needed to establish a structured-data norm when the crawl has fewer than 10 pages.
• Skipped with the message No page type has 8+ comparable pages sharing a structured-data pattern when no page-type group reaches 8 pages with an agreed set of types.
• Skipped with the message No pages available for analysis when the crawl produced no pages at all. This wording comes from the non-streaming path; a streaming crawl with no pages takes the Only 0 page(s) crawled branch instead.
• Passes with the message All N comparable page(s) carry the structured data their page type establishes (article Article, product BreadcrumbList+Product) when every judged page is inside its group’s norm. The parenthesised summary lists up to 10 groups, sorted by page type, each as the lowercased page type followed by its norm types in alphabetical order joined with +.
• Warns, or fails, with the message N of M page(s) are missing structured data their same-type siblings have (article Article, product BreadcrumbList+Product). Each item reads 40 of 45 product pages have Product schema, these 5 do not and carries up to 20 example URLs. The status is fail when the deviant pages are 20% or more of the judged pages, and warn below that.

A schema @type joins a group’s norm when at least 70% of the group carries it, and pages below that share establish no norm at all, so a site whose same-type pages genuinely differ produces no finding. A BreadcrumbList present on 70% of a group does become part of that group’s norm, and the 30% without it are reported. Norms are computed per @type rather than by exact type set, so one optional extra cannot shatter the majority and hide a real gap behind it. Items are capped at 10 gaps and the norm summary at 10 groups.

Grouping is by page type, never by the rich result types derived from the markup. Deriving the group from the field being judged would put a product page missing its Product markup in a different group from the pages it should be compared against, where it could never be an outlier. The rich-result flag is still recorded per gap, since a missing rich-result-eligible type costs a real search feature.

How to fix it

Find where the passing pages get their JSON-LD, then make sure the listed pages run through the same code path. The gap is usually one template branch, one content type in the CMS, or a set of older pages created before the markup existed.

Pages built from one template are safe by construction: they all carry the same types, so they sit inside their own norm and none of them can be a deviant.

A page that genuinely should not carry the markup, such as a discontinued product stub or a landing page filed under the same type, is fine to leave. The finding says only that it is currently the exception.

Enable / disable

Disable this rule

[rules]
disable = ["schema/coverage-outlier"]

Disable all Structured Data rules

[rules]
disable = ["schema/*"]

Enable only this rule

[rules]
enable = ["schema/coverage-outlier"]

enable replaces the default ["*"], so this is the only rule selected. Leave disable out: a rule that matches any disable pattern is skipped even when enable names it. Only an enabled setting under a rule’s own [rule_options] table overrides both lists.

Related rules

• schema/product: the per-page check for the type this rule most often finds missing.
• schema/article: the same for article and blog pages.
• content/title-pattern-outlier: the same site-norm approach applied to page titles.
• content/thin-vs-site-norm: pages far shorter than the site’s own norm for their page type.

Structured data findings ship in every audit next to the core SEO, performance and agent experience rules. See Fix SEO issues with AI for how an agent works through a report.

References

• Introduction to structured data markup, Google Search Central
• Search gallery of rich result types, Google Search Central
• Getting started with schema.org, schema.org

Check your site

Run squirrel audit https://example.com and open the Structured Data section of the report. Each gap lists the page type, the missing type and the URLs. Signed-out local audits need no account and cost nothing. Signed-in audits spend credits.