Owner, trigger and scope
The SEO lead owns the procedure: they read every new issue, decide whether it matters, and own the ticket until it is validated. The developer owns fixes to templates, server configuration, redirects and page performance. Content and metadata fixes go to the web publisher through the normal publishing procedure.
There are two triggers. Weekly, the SEO lead reviews the indexing, experience and enhancement reports for new or growing issues. Ad hoc, an alert email from GSCGoogle's search performance and indexing tool. triggers the procedure the same working day, because alerts are usually sent when an issue is new or rising.
The scope covers issues surfaced in Search Console reports: page indexing, sitemaps, experience reports, structured data enhancements, manual actions and security issues. Manual actions and security issues are escalated immediately to the head of marketing and the technical lead as well. Broader periodic reviews of site health belong to the technical SEO audit; this SOP handles issues as they arrive between audits.
Before you start: inputs and access
| Input | Owner | Notes |
|---|---|---|
| Search Console property access | SEO lead | Full access is needed to validate fixes |
| Issue log | SEO lead | Every triaged issue, its decision and status |
| List of intended noindex and canonicalised pages | SEO lead | Separates expected states from problems |
| Release and change history | Developer | Shows what changed near the issue's start |
| Crawler or URL inspection tool | SEO lead | Reproduces what search engines see |
| Ticket system access | Developer | Where fixes are assigned and tracked |
The Google Search Console guide explains each report and what its states mean. Keep it open during triage; this SOP assumes you can read the reports and focuses on what to do with them.
The Page indexing report is where most triage starts, and its status names look alike but call for different responses. The table gives the usual reading of the ones met most often, using the wording Search Console shows at the time of writing. Google renames and adds states, so confirm the current wording and meaning in the Page indexing report help before relying on a row. "Usually" matters: whether a status is expected or a problem for a given URL still depends on whether that page is meant to be indexed, which the list of intended exclusions settles.
| Status in the report | Usually expected, or a problem | First check | Usual owner |
|---|---|---|---|
| Excluded by 'noindex' tag | Expected where the page is on the intended-noindex list | Is the URL on the list? If not, find what added the tag | SEO lead, then developer |
| Page with redirect | Expected for old URLs that redirect on purpose | Does the redirect land on the right page, and do internal links or the sitemap still list the old URL? | SEO lead, developer |
| Blocked by robots.txt | Expected for sections deliberately kept out of crawling; a problem for pages meant to rank | Which robots.txt rule matches the URL, and was it meant to? | Developer |
| Duplicate without user-selected canonical | Usually a problem to review | Which URL does Google treat as the main one, and should the page declare its own canonical? | SEO lead, developer |
| Soft 404 | Usually a problem | Does the page read as empty or "not found" while loading normally? Fix the content, or return a real not-found status | Developer, web publisher |
| Server error (5xx) | A problem | Does the URL fail now, and do the server logs show errors when Google crawled it? | Developer |
| Crawled - currently not indexed | A problem for pages meant to rank; often acceptable for thin or low-value pages | Is the page distinct and useful, and linked from elsewhere on the site? | Web publisher, SEO lead |
| Discovered - currently not indexed | A problem if valued pages stay here for long | Is the page linked internally, and is the server answering crawl requests reliably? | SEO lead, developer |
The procedure
Read the alert
Open the report the alert or weekly review points to and note the issue type, the first detected date and the trend. Read the exact wording of the reason, since similar-sounding states have different causes, and match it against the status table above. Check the list of intended exclusions: a noindexedA robots directive telling search engines not to show a page in their results. archive or a correctly canonicalised duplicate reported as excluded is expected, and is logged as such and closed without further work.
Size the affected pages
Export the example URLs and work out which templates or sections they belong to. Decide how much the issue matters: are the affected pages ones that should rank and earn traffic, or low-value pages? Check their recent clicks in the performance report. An issue affecting a handful of thin pagesPages that offer little original value of their own, such as near-empty templates, near-duplicates or copies with only a word or place changed. is low priority; one affecting a product template is urgent, whatever the count.
Reproduce the issue
Use URL inspection on several example URLs and test the live URL to see what Google's crawler sees now. Crawl the same URLs with your own crawler and load them in a browser. If the issue no longer reproduces, it may already be fixed or intermittent; note which, and check server logs for intermittent errors before assuming it has gone away. For server errors on pages that load normally in a browser, the page on server errors in Search Console shows how to group the examples by template and time.
Find the cause
Trace the issue to its source: a template change, a redirect rule, a robots rule, a Canonical TagTells search engines the preferred URL. pointing at the wrong URL, a sitemap listing removed pages, slow server responses or a structured data field missing from a feed. Compare the first detected date with the release history. The crawling and indexing guide explains how each mechanism works, including when Crawl BudgetHow much a search engine crawls a site. genuinely matters.
Assign the fix
Write a ticket for the owner of the cause with the issue, the affected templates, example URLs, the evidence and the expected result once fixed. Set a priority from the sizing step. For experience issues, the Core Web Vitals guide explains which template-level fixes tend to move many URLs at once. Agree a target date with the developer and record it in the issue log.
Validate in Search Console
Once the fix is deployed, test several example URLs live with URL inspection to confirm the issue is gone, then start validation for that issue in Search Console. The Search Console Help documents validation as a recheck of the affected URLs that takes time to complete; its Page indexing report help is the place to check the current wording. Watch its progress; if it fails, read which URLs still show the issue and return to the cause step with those examples.
Log and close
When validation passes, or the issue count falls to the level expected, record the outcome in the issue log: cause, fix, ticket, dates and the validation result. Close the ticket. If the issue came from a release that lacked an SEO check, note it and raise a change to the release process so the same cause is caught before it ships next time.
A redirect rule that removed category pages from the index
Anvilwhistle Hardware, an invented tools retailer, receives a Search Console alert about pages with redirects. Sizing shows 1,800 affected URLs, including 42 category pages that had earned around 30% of organic clicks. Reproducing the issue shows category URLs with a trailing slash now redirect to a search results page, a rule introduced in a release 9 days earlier. The developer corrects the rule the same day, the SEO lead checks 10 examples live and starts validation, and validation passes 3 weeks later. The issue log records a new step adding redirect checks to every release.
When the alert concerns pages with redirects, the page on page with redirect in Search Console separates the redirects working as intended from the links, sitemaps, canonicals and rules that contradict them.
QA checks
An issue that cannot meet every check stays open in the log with a note of what is missing, rather than being closed on the assumption that it will resolve itself.
Before closing an issue, the SEO lead confirms:
- Issue classified as expected or as a problem, with the reason recorded.
- Affected pages sized by template and by their value, not only by count.
- Issue reproduced on live URLs, or noted as intermittent with evidence.
- Cause identified and linked to a change where one exists.
- Fix deployed and checked live on several example URLs before validation.
- Validation passed in Search Console, or a documented reason why it is still pending.
- Issue log updated with cause, fix, ticket and dates.
- Escalation made for any manual action or security issue.
Hand-offs and records
The SEO lead hands each confirmed problem to its owner through a ticket and remains responsible until validation passes. Developers hand back with the deployed change and its release reference. Content fixes go to the web publisher and follow the publishing procedure. Manual actions and security issues are handed to the head of marketing and the technical lead immediately, with the report details, because they can affect the whole site.
The issue log is the main record, and it is the first input to the next technical audit: recurring causes there show where the release process or templates need a permanent change. Keep the log alongside the Search Console annotations or your own change log, so a later traffic change can be matched to the issue that caused it.
When a row turns on whether a URL should be blocked from crawling or kept out of the index, the comparison of noindex and robots.txt disallow gives the rule for choosing between them.
Common mistakes
Most stem from treating Search Console as a to-do list instead of a set of signals to interpret.
- Fixing expected states. A deliberate noindex reported as excluded is working as intended. Log and move on.
- Prioritising by count alone. A few valuable templates matter more than many thin pages.
- Validating before checking live. Validation that fails restarts the wait. Test examples first.
- Treating the symptom. Removing a URL from a sitemap hides an issue without fixing its cause.
- No release link. Issues that started with a release recur unless the release process changes.
- Ignoring alerts until the weekly review. Alerts usually signal something new or growing. Read them the same day.
- Closing on a falling count. A shrinking issue may still affect valuable pages. Check the examples.
