Playbook: triage 404s Googlebot actually requests

Playbook: triage 404s Googlebot actually requests
Four sources for fixing the 404s that matter, ignoring the noise.
→ Google's "are 404s bad" doc — most are harmless; ignore them by default. Takeaway: only act on 404s with internal links or bot interest, per the docs.
→ awk recipe (Server Fault) — filter to $9==404 and Googlebot, then uniq -c by URL. Takeaway: rank by request count, fix the top of the list only.
★ Pick of the week — Aleyda Solis' 404-prioritization note — a 404 that used to be a 200 and still gets links is the urgent one. Takeaway: cross-reference against your old sitemap to find the regressions.
→ Search Console's crawl-stats by response — confirms the trend at the host level. Takeaway: a rising 404 line means a new bug, not legacy cruft.
Volume plus links equals priority — a 404 nobody requests isn't a problem.