Crawl Error vs 404 Error: The Core Distinction
A crawl error is any condition that prevents a search engine bot from successfully accessing and processing a URL — this includes DNS failures, server timeouts, robots.txt blocks, and HTTP error responses. A 404 error is one specific HTTP status code (404 Not Found) that a server returns when the requested resource does not exist. Every 404 is a crawl error, but not every crawl error is a 404.
The practical distinction matters because the fix differs. A 404 tells Google the page is gone — the server responded, it just has nothing to return. A broader crawl error may mean the server never responded at all, the DNS lookup failed, or a redirect loop consumed the bot's request before any status code was served. Diagnosing the wrong one wastes engineering time and leaves SEO damage unresolved.
How Each One Works Mechanically
When Googlebot requests a URL, it follows a chain: DNS resolution → TCP connection → HTTP request → server response. A 404 error occurs at the final step — the server is reachable, processes the request, and deliberately returns status code 404. The page simply does not exist at that path. The bot receives a valid, complete HTTP response; it just carries bad news.
A crawl error can abort at any earlier step. If DNS resolution fails, the bot never opens a connection. If the server times out, no response arrives. If a redirect chain loops indefinitely, the bot gives up before receiving any status code. Google Search Console labels these differently: 404s appear under 'Not Found,' while server errors, redirect errors, and DNS errors each get their own category under the broader crawl error umbrella.
For an ecommerce store, this means a deleted product page typically generates a 404, while a misconfigured CDN or an overloaded server during a flash sale generates other crawl error types. Both show up in Search Console, but under different diagnostic buckets.
When Each Condition Applies in Ecommerce
404 errors are the most common crawl issue for ecommerce sites because product catalogs change constantly. A SKU is discontinued, its URL is deleted, and any internal link or backlink pointing to that URL now returns 404. Seasonal landing pages left live without content, variant URLs for out-of-stock sizes, and paginated category pages that shrink are all routine 404 sources.
Non-404 crawl errors appear in different scenarios. A site migration that breaks DNS propagation causes connection-level crawl errors across the entire domain, not just specific pages. A poorly written redirect rule creates loops that bots cannot exit. A server configured to return 500 Internal Server Error for deleted pages instead of 404 generates server errors in Search Console — a worse signal, because Google treats 500s as temporary problems and keeps recrawling, consuming crawl budget.
Understanding which condition applies changes the triage priority. Widespread non-404 crawl errors indicate infrastructure problems that affect all pages. A spike in 404s indicates a content management issue that affects specific URLs. Both damage crawl efficiency, but in different ways and at different scopes.
Where They Overlap and How They Interact
The overlap zone is this: a 404 is a successful HTTP transaction that reports a missing resource, while a crawl error is any failed transaction — including 404s. Google Search Console separates them in its interface, but in the underlying crawl log, a 404 is recorded as a crawl event with a known outcome. The bot did its job; the content was absent.
They interact most visibly during site migrations. When a store moves from one URL structure to another without setting up redirects, thousands of old URLs return 404s. Those 404s are crawl errors in the broad sense, but they are not infrastructure failures — they are addressable with 301 redirects. If the migration also introduces server misconfiguration, the same crawl report contains both true 404s and separate server-timeout errors, and they must be resolved through different workflows.
A common misread is treating all red flags in Search Console as '404 issues.' Non-404 crawl errors left unresolved will cause Google to reduce crawl frequency on the domain, which delays indexing of new and updated pages — a direct revenue risk for stores publishing new inventory regularly.
Actionable: How to Triage and Resolve Each Type
Start in Google Search Console under Indexing → Pages. Separate the 'Not Found (404)' entries from all other error types. For 404s: identify which carry inbound backlinks or internal links using a crawl tool, then decide between three responses — 301 redirect to the nearest relevant live page, restore the page if the content still has value, or accept the 404 and clean up internal links pointing to it. Soft 404s (pages that return 200 but display 'product unavailable' content) need a separate pass because Search Console flags them distinctly.
For non-404 crawl errors, categorize by type. DNS errors require coordination with your hosting or domain registrar. Server errors (5xx) require server-side debugging, often involving your DevOps team or hosting support. Redirect errors require auditing your redirect rules for loops or chains longer than three hops. Fix infrastructure errors before optimizing 404 handling — a server that randomly returns 500s will contaminate any crawl budget analysis.