How to Fix 404 Errors in Google Search Console

What "Not Found (404)" in Search Console Actually Means

A "Not found (404)" entry in the Page Indexing report means Googlebot requested a URL and your server responded that nothing lives there. It's not a penalty, not a ranking signal on its own, and not proof something is broken — just a record that a URL, at crawl time, returned an HTTP 404 status.

The confusion starts because this line sits inside the "Why pages aren't indexed" section of the Google Search Console page indexing report, next to genuinely worrying entries like server errors or redirect issues. Grouped together, they look equally urgent, but they aren't. Google has been explicit that 404s are a normal, expected part of how the web works — pages get deleted, products go out of stock, campaigns end, and Googlebot eventually notices and stops trying to index them.

The reframe worth internalizing before you touch anything: most 404s in the not found 404 Google Search Console list are Google doing exactly what it's supposed to do. Your job isn't to clear the list to zero — it's to find the small subset of 404s that actually cost you traffic, links, or rankings, and leave the rest alone.

How to Find Your 404 Errors in Google Search Console

The current path is: Indexing > Pages, then scroll to "Why pages aren't indexed" and click Not found (404). That opens the full list of URLs Google has crawled and found to return a 404, along with the number of affected pages and a sample date range.

Export the list (download option, top-right of the table) before you triage anything — it gives you a stable working copy instead of re-navigating GSC for every check. You'll also see "Submitted URL not found (404)," a stricter variant meaning you (or your sitemap, or an internal link) explicitly submitted that URL for indexing and it still 404s. These deserve slightly more attention, since something on your own site pointed Google there on purpose.

Click any individual URL and use the URL Inspection tool to confirm the live status, see the last crawl date, and check whether Google is treating it as indexed or excluded. This is also where you'd request a recrawl once you've applied a fix.

If you're new to Search Console generally, it's worth spending time on the platform's other reports too — see our Google Search Console tips that actually drive action for guidance beyond the indexing report.

Not All 404s Need Fixing: The Triage Test

Before you fix 404 errors one by one, run each URL through a three-question test:

1. Does it have backlinks? Check your backlink tool (or GSC's Links report as a baseline) for external sites linking to that exact URL.
2. Does it have internal links? A crawl of your own site shows whether old navigation, blog posts, or sitemaps still point to the dead page.
3. Did it have historical traffic or rankings? Check Search Console's Performance report, filtered by that URL, for past impressions or clicks.

If the answer to all three is no, the URL is safe to leave as a 404. It cost you nothing when it existed, and it costs you nothing now. This is the single biggest thing most guides get wrong about 404 errors SEO impact: they treat every row as lost equity, when a 404 on a URL nobody linked to and nobody visited is just Google's index catching up with reality.

If any answer is yes, the page had value worth preserving, and you move to deciding which of the four fixes applies. This triage step is what separates a ten-minute cleanup from a project that eats your afternoon for no measurable gain.

The Four Fixes: Ignore, 410, 301 Redirect, or Restore

Once you know which URLs matter, each one gets exactly one of four treatments.

Ignore. Use this when the triage test comes back negative on all three counts — no backlinks, no internal links, no traffic history. Leaving it as a 404 is correct; Google will drop it from the index on its own over time.

410 Gone. Use this when the page is permanently removed and you want to signal that clearly and immediately, rather than letting Google figure it out through repeated crawls. A 410 tells Googlebot "this is gone on purpose, stop checking," which can help it deprioritize the URL faster than a plain 404. It's the right call for discontinued products or expired content with no replacement and no meaningful backlinks to preserve.

301 redirect. Use this when the page had backlinks, internal links, or traffic, and there's a genuinely equivalent replacement — a merged article, an updated product, a renamed category. A 301 passes most historical authority forward and keeps visitors landing somewhere relevant instead of on a dead end.

Restore. Use this when the 404 is actually a mistake — a page deleted accidentally, a URL changed unintentionally during a template update, or content pulled down that shouldn't have been. Bring the original URL back rather than redirecting or replacing it.

The trap most teams fall into is treating the homepage as a universal 301 target. Redirecting 404 to homepage feels safe, but Google treats a mass pattern of unrelated URLs redirecting to an unrelated destination as a signal that the redirect isn't meaningful — functionally the same problem as a soft 404, and it dilutes the value a real 301 redirect vs 410 decision was supposed to preserve. If you can't find a genuinely relevant replacement page, a 410 or a plain 404 is the more honest signal than redirecting to the homepage.

404 vs. Soft 404: Don't Mix Up the Fix

A regular 404 and a Soft 404 look similar in the report but describe opposite technical situations, and applying a 404 fix to a soft 404 problem (or vice versa) won't work.

A standard 404 means the server correctly returned a 404 HTTP status code — Google trusts the signal and treats the page as genuinely gone. A Soft 404 means the server returned a 200 OK status for a page Google believes is actually empty, broken, or too thin to be useful — an out-of-stock page with no content, an error message rendered inside a normal template, or a JavaScript-rendered page that fails silently.

The soft 404 vs 404 distinction matters because the fix is different. A true 404 usually just needs triage per the process above. A soft 404 needs a server-side fix: either make the page return a real 404/410 status if it should be gone, or add substantive content if it should stay live. Redirecting a soft 404, or ignoring it the way you'd ignore a harmless true 404, leaves the underlying problem — a page telling Google "I'm fine" while providing nothing of value — completely unaddressed. If you're regularly finding mismatches between what a page returns and what it actually contains, that's often tangled up with broader indexing and canonicalization issues; see canonical tag troubleshooting: 5 failures basic audits miss for the adjacent failure modes.

Keeping 404s From Piling Up Again

Prevention is cheaper than cleanup. Three habits stop most future 404 problems before they reach the report:

• Update internal links whenever you delete or move a page. A quick site search for the old URL before you take it down tells you exactly which navigation menus, blog posts, or sitemaps need updating.
• Check backlinks before removing any URL, not after. If external sites link to it, that's your signal to redirect rather than delete outright.
• Re-crawl after migrations. URL structures change, and even a well-planned move often generates a temporary spike in 404s if redirects weren't mapped for every old path. Our SEO site migration checklist covers how to plan redirects before, not after, a move happens.

404s are one entry in a much longer list of technical issues that compete for your attention — see how to fix technical SEO issues: a priority action plan for how to weigh them against other fixes.

The real limitation of the process above is that it's manual. Checking the Page Indexing report every few weeks means new 404s sit unnoticed between checks, and by the time you catch one with real backlinks, you may have already lost the ranking signal it was carrying. Monitor 404 errors continuously rather than periodically, and you catch the ones that matter — new backlink-carrying 404s, migration fallout, accidental deletions — within hours instead of weeks. That's the case for automated, ongoing detection over manual spreadsheet exports, a shift covered in more depth in site audit tool: what it checks and how to use one.

Manual triage works fine for a one-time cleanup on a small site. It stops scaling the moment you're publishing regularly, running a large site with years of URL history, or managing several properties at once — at that point, checking GSC every few weeks isn't a process, it's a gap. Rankevra automates the whole workflow: detecting new 404s as they appear, running the backlink-and-traffic triage automatically, distinguishing real regressions from harmless noise, and queuing the right fix — redirect, 410, or restore — instead of leaving it for someone to eyeball a spreadsheet. If you'd rather not re-run this checklist by hand every month, Rankevra handles the detection and the fix in one workflow.

Frequently Asked Questions

Do I need to fix every 404 error in Google Search Console?

No. Most 404s in the Page Indexing report are pages that no longer exist and shouldn't — deleted products, expired campaigns, old test pages. Only URLs with backlinks, internal links, or historical traffic need action; the rest are safe to leave as-is.

What's the difference between a 404 error and a soft 404?

A true 404 returns an actual 404 HTTP status code, telling Google the page is genuinely gone. A soft 404 returns a 200 OK status — the server says the page loaded successfully — even though the content is empty, broken, or too thin to be useful, so Google treats it as a de facto 404 anyway.

Should I redirect all my 404 pages to the homepage?

No. Redirecting unrelated deleted pages to the homepage doesn't preserve relevance, and Google can treat that pattern the same way it treats a soft 404. Only redirect to a page that's a genuine, topically relevant replacement for the one that's gone.

How long do 404 errors stay in the Search Console report?

404s typically remain listed until Google re-crawls the URL, confirms it's still gone, and gradually ages the entry out of the report over subsequent crawls. There's no fixed removal date — it depends on how often Googlebot revisits the URL.

What's the difference between a 404 and a 410 status code?

A 404 says a page isn't found, without specifying whether that's temporary or permanent. A 410 explicitly says the page is gone permanently and isn't coming back, which can prompt Google to deprioritize re-crawling it sooner.

Can 404 errors hurt my SEO rankings?

Only if the 404'd URL previously earned backlinks, rankings, or traffic that isn't redirected anywhere. A 404 on a page nobody linked to or visited has no meaningful SEO impact.