Why this matters
Crawlers and GSCGoogle's search performance and indexing tool. report issues at alarming volumes and mostly meaningless severities: a deliberate noindex on a basket page sits beside a robots rule hiding the pages you want to rank. Triage means reading each line as a symptom, tracing it to a cause and judging the cause by the pages it affects. The crawling and indexing guide explains every mechanism in this exercise, and the Search Console error triage procedure gives the read, size, reproduce and assign sequence you practise here.
The crawl and index reports
Pageturn Books is an invented independent bookshop selling books online. Its owner exported both reports after an audit tool emailed a red warning. The shop has one part-time developer.
Pageturn Books: combined issue list
| Ref | Source | Issue as reported | Affected URLs | Example URL | Notes from the owner |
|---|---|---|---|---|---|
| A | Search Console | Excluded by noindex tag | 1,240 | /search?q=tolkien | Includes /basket and /account |
| B | Search Console | Alternate page with proper canonical tag | 3,900 | /fiction?sort=price | Sort and filter views |
| C | Search Console | Duplicate, Google chose different canonical than user | 46 | /fiction/crime/ | Category pages; nav links use the trailing slash, while the canonical tags declare the version without it |
| D | Search Console | Blocked by robots.txt | 180 | /reading-guides/book-club-picks | Guides are in the sitemap; new this year |
| E | Search Console | Discovered, currently not indexed | 610 | /books/the-salt-path-9781... | Mostly titles added in the last two months; they are linked only from deep pages of the paginated category lists |
| F | Search Console | Crawled, currently not indexed | 2,050 | /author/jane-smith | Author tag pages, most list one book |
| G | Search Console | Not found (404) | 870 | /books/out-of-print-title | Removed listings; 35 have external links |
| H | Search Console | Soft 404 | 22 | /non-fiction/falconry | Empty categories showing "no books found" |
| I | Crawler | Internal links to redirects (chains) | 4,300 links | old http:// links in blog posts | Two hops each |
| J | Audit tool | "Crawl budget wasted on parameter URLs" | n/a | — | Red warning; the shop has about 9,000 book pages |
| K | Crawler | Sitemap lists URLs returning 404 | 540 | /books/out-of-print-title | Sitemap is never regenerated |
Clicks in the last three months, by section: category pages 41%, book pages 38%, reading guides 0% (blocked), blog 12%, author pages 2%, other 7%.
Grouping issues by cause
Eleven rows are not eleven problems. Ask of each row what setting, template or process produced it; rows that share a cause share a fix. Size each cause by the pages it touches, not the count beside it: a few category URLs earning many clicks outrank thousands of author pages earning almost none. Then separate expected states from faults. A page excluded on purpose is the report confirming that a control works.
Read the Canonical TagTells search engines the preferred URL. rows with care: the same category can mean "working as intended" or "Google disagrees with you", depending on which URL links, sitemap and canonical favour.
Task
- Read every row and write, in one line, the cause you think produced it.
- Merge rows that share a cause into groups, and name each group.
- For each group, note the pages affected and how valuable they are, using the clicks by section.
- Mark each group fix now, fix later or ignore, with a one-sentence reason.
- Write the fix order for the developer: what to change, on which template or file, and how to confirm it.
- Note anything you would check before accepting the audit tool's red warning.
Worked answer
Group one: a robots rule hiding the reading guides (row D). Fix now. The guides are listed in the sitemap and blocked from fetching: "index these" and "do not read these" at once, probably a disallow rule left from staging. The pages cannot be read, so they rank poorly if at all. Remove the rule, test examples live in URL inspection, then validate.
Group two: category canonical conflict (row C). Fix now. Navigation links to the trailing-slash version while the canonical declares the version without it, on the shop's largest source of clicks. Redirect the variant and update the navigation template so links, sitemap and canonical agree; the site architecture guide explains why navigation links weigh so heavily.
Group three: new titles not yet crawled (row E). Fix now. Recent books sit deep in pagination. This is discovery, not quality: link new titles from a new-releases hub in the main navigation. The sitemap side of discovery is handled once, in group four.
Group four: removed listings and a stale sitemap (rows G and K). Fix later. Most not-found pages are correct: the books are gone. Redirect only the listings with external links, and only where a genuine equivalent exists (another edition, or the author's page if it has substance). Regenerate the sitemap automatically so it lists canonical, indexable URLs only, new titles included.
Group five: thin and empty templates (rows F and H). Fix later. Single-book author pages were judged not worth storing; empty categories read as missing pages. Return a not-found status for empty categories, and enrich author pages for prolific authors or noindex the rest.
Group six: link hygiene (row I). Fix later. Update the old blog links to final URLs in one batch.
Ignore with a reason: rows A, B and J. Search, basket and account pages are noindexed on purpose. Sort and filter views pointing at their canonical are the control working. And Crawl BudgetHow much a search engine crawls a site. is a concern for very large, fast-changing sites; a shop of this size is not limited by it, so the red warning describes parameter URLs that row B already shows are handled.
Fix order for the developer: the robots rule, then the category redirects and navigation links, then the new-releases hub, then the redirects for linked removed listings and the automatic sitemap regeneration, then empty categories and author pages, then the blog links. Each ticket names example URLs and the report where the fix will show. Page-speed problems go to a separate queue judged with the Core Web Vitals guide.
Common mistakes
- Sorting by count. The biggest numbers on the list (the 3,900 sort and filter URLs in row B and the 4,300 redirecting links in row I) sit in the ignore and fix-later groups, and the 2,050 author pages in row F are near the bottom of the real order.
- Fixing expected states. Removing noindex from search results pages would create problems, not solve one.
- Redirecting every removed book to the homepage. It is treated as a soft not-found and wastes the effort.
- Requesting indexing for thin pages. The judgement was about quality; the request does not change it.
Compare your groups with the ones above, then return to the intermediate SEO level and take the technical SEO checkpoint.
