Research Guides
Test the Breadcrumb Trail After a Content Reorganization
Moving an article into a new category can leave a small but confusing remnant: a breadcrumb that still sends readers to the old section. The article loads, the new menu looks right, and the route above its title quietly tells a different story.
This is a focused editorial and navigation test for an existing site. It does not claim that Colquhoun.shop offers a breadcrumb audit tool or that adding markup guarantees a search appearance.
1. Write the intended hierarchy first
Record the article, its intended parent section and the broader category, if one exists. Describe why each parent is useful to a reader. Do not infer the hierarchy solely by splitting the URL at slashes or copying the reader's last few clicks.
For an invented research library, an article on comparing subscription limits might belong under “Research Guides” and then “Pricing Research.” If “Pricing Research” is only a tag with no usable destination, identify that before making it a clickable parent in the trail.
2. Follow every parent link from the article
Open the article directly, as someone arriving from a shared link would. Activate each breadcrumb destination and confirm that its title and content match the label. A successful response is not enough if “Pricing Research” opens an unrelated general archive.
Repeat the route with a keyboard and check the trail at a narrow viewport. Record whether the current page is distinguishable and whether long labels remain usable. These are focused observations, not a complete accessibility evaluation.
3. Compare the structured trail with the actual route
Google describes breadcrumbs as a way to indicate a page's position in a site's hierarchy. Its documentation supports multiple breadcrumb trails when there are multiple navigation paths, and recommends validating structured data before checking deployed pages. Google Search Central: Breadcrumb structured data
If the page includes breadcrumb markup, compare its labels, destination URLs and ordered positions with the intended hierarchy. Inspect all generated trails, not only the one most obvious on screen. An old trail can remain in a shared template even after the visible navigation changes.
Keep the validation result separate from the navigation result. A parser can accept structured data without proving that the linked parent contains the promised material. Likewise, a working visible trail does not prove that the deployed structured data describes it correctly.
4. Retest the shared component, not just one article
Choose a small, explicit sample: the moved article, a sibling in its new section, an article remaining in the old section and a page with a longer trail. Include an article with more than one legitimate classification if the site supports that.
Look for hard-coded category names, missing parents and duplicate positions. If only four pages were checked, say so. Do not turn that sample into a claim that every trail on a large site has been verified.
5. Save a route-level acceptance record
- Page: exact article URL and observed deployment or check time.
- Intended trail: labels, order and reason for each parent.
- Live route: actual destinations and matching content.
- Markup: every structured trail inspected and validation result.
- Coverage: sample pages, keyboard and viewport checks.
- Outcome: scoped repair, retest or unresolved mismatch.
For broader markup review, pair this with the structured-data QA worksheet. The immediate goal is straightforward: a reader who climbs from an article to its parent section reaches a coherent place, and the machine-readable description does not contradict that route.