Measurement / framework / Free to use
Search Indexing Checklist: Fix Defects Without Chasing Every Exclusion
A URL-level worksheet separating intended pages, legitimate variants, missing evidence and verified fixes before requesting another crawl.
Growthcraft Editorial · 2026-10-27. AI-assisted research and implementation. Examples are synthetic; Akshay's personal review is not claimed.
Copyable template
Select and copy the complete template below. With JavaScript enabled, you can edit, copy and download it in the interactive workspace.
INDEXING EVIDENCE TRIAGE — original Growthcraft method Decision and business owner: Engine/property, scope, inventory version and observation cutoff: Desired canonical URLs (one unique row each): Separate variant list and expected destination: Engine status source, extraction time and observation date per URL: Status: indexed / confirmed not indexed / unknown. Never infer a status from an omitted example. Technical evidence: HTTP status, final URL, robots access, meta/header directives, declared canonical, rendered content, referring link, sitemap membership. Engine-selected canonical and last crawl where available: Commercial priority and intent: what useful task does this page solve? Disposition: no change / repair verified defect / improve usefulness and links / inspect missing evidence / investigate host response. Fix owner, exact scope, evidence before/after and rollback: Submission state: not attempted / confirmed requested / failed / uncertain. A click is not confirmation. Follow-up: observation date, indexed status, impressions, clicks and genuine enquiries separately. Approval: never remove a useful page solely to make a percentage look better.
A decision for each URL
Use this original framework when a growing content library has mixed index statuses. The deliverable is a scoped decision log, not a promise to make every report row green. Begin with the pages that support an actual customer task.
Four gates
- Intent: identify the desired canonical page and keep aliases separately. Do not call a correctly redirected HTTP URL a missing landing page.
- Evidence: retain engine, observation date and source per URL. An omitted example is unknown. A successful live fetch answers a different question from stored index status.
- Action: fix a proven accidental directive or broken response; otherwise review usefulness, overlap and contextual discovery. Assign an owner and smallest reversible change.
- Outcome: record deployed, requested and indexed independently; evaluate traffic and enquiries in complete comparable windows.
Worked synthetic review
A 100-page intended inventory contains 65 confirmed indexed, 20 confirmed not indexed and 15 unclassified pages. Thirty additional redirect variants belong outside that inventory. The current confirmed share is 65%; observation coverage 85%. The missing-status range is 65–80%, not a forecast. Use the coverage calculator to verify the arithmetic.
Useful handoff
Give the developer a specific URL, captured response, unintended setting, proposed repair and repeatable check. Give the editor the page purpose, competing internal content and a useful next reading path. If all you have is a category label, assign investigation rather than a fake fix.
When to stop
Do not mass-remove content, add noindex to reduce the excluded count, or mark an entire category fixed without evidence. A search engine can decline technically accessible pages. Keep unknowns visible and commercial priorities separate from arithmetic coverage.
Sources and boundaries
Checked October 27, 2026. Google Page indexing documentation distinguishes legitimate exclusions and limits on example lists. Bing's URL Inspection introduction (September 2020) distinguishes indexed evidence and live checks. These sources support the diagnostic distinction, not this original worksheet or its synthetic numbers. No engine access, submission or indexing promise is included.