Original framework
Six-stage indexation decision tree
Start at discovery and move forward only when the previous stage has evidence of working.
Start with the exact URL and observed status
Record the canonical URL, the Search Console Page indexing status where available, the last crawl date and whether the problem affects one page or a repeatable template. Do not diagnose from a site search alone; it is not a complete index report.
1. Discovery: can the page be found?
A page that appears only in an XML sitemap has a weaker relationship to the site than one linked from relevant navigation, service hubs and contextual content. Confirm that crawlable HTML links point to the preferred URL.
2. Access: can the page be crawled?
Test the production response without a browser session. Review status codes, redirects, robots rules, authentication, firewall challenges and user-agent differences. A page can work for a person while a legitimate crawler receives a block or challenge.
3. Rendering: is the useful content present?
Inspect initial and rendered HTML. Important headings, service details and crawlable links should not depend on an interaction that a crawler may never perform. Check whether scripts fail, resources are blocked or placeholders remain in rendered output.
4. Consolidation: is another URL winning?
A canonical tag is one signal, not a command. Redirects, internal links, sitemap entries, duplicated content and external references should consistently support the preferred URL. Investigate why another version may be a more coherent canonical candidate.
5. Value: is the page genuinely distinct?
Service and location pages often fail when they repeat the same text with minor substitutions. Clarify the service, audience, operating reality, proof, decisions and conversion path that make the page useful. Consolidation may be better than expanding a thin set.
6. Processing: has the change been evaluated?
After a verified fix, update internal signals, submit an accurate sitemap and use URL Inspection appropriately. Recrawling and processing take time, and an indexing request cannot override unresolved access, canonical or quality problems.
What not to do
Do not remove canonical tags indiscriminately, publish more duplicate pages, repeatedly request indexing, or assume a sitemap guarantees inclusion. Preserve evidence before changing several signals at once so the eventual cause can be understood.
Primary sources
Platform behavior and documentation can change. These primary references were reviewed for this guide: