What Is a 404 Error, and How Do Search Engines Handle It?

A 404 error happens when someone — a person or a search engine crawler — requests a URL and the server responds that nothing exists at that address. It's one of the most common status codes on the web, and also one of the most misunderstood: a lot of business owners treat every 404 as damage that needs immediate repair, when in reality a normal number of them is completely expected. The useful skill isn't eliminating 404s entirely — it's knowing which ones actually cost you something and which ones don't.

What technically happens during a 404

"404" is an HTTP status code — the three-digit number a web server sends back along with every response, indicating what happened with the request. A 200 means "here's the page, successfully." A 301 means "this moved permanently, here's the new address." A 404 means "Not Found" — the server has no content to return at that URL. Most CMSs and hosting platforms display a generic (or custom-designed) "page not found" message to visitors when this happens, but the status code itself — the part search engines and browsers actually read — is what determines how the request gets handled and recorded.

Common causes of a 404

• A page was deleted without a redirect. The content was removed and nothing was set up to send visitors somewhere else.
• A typo in a URL or a link. Either in a link on your own site or one written on an external site, a single mistyped character produces a dead link.
• A broken external link. Other sites link to a URL of yours that has since changed or been removed, and they never update their link.
• Content moved without updating the old URL. A page was restructured or renamed, but the old address was never redirected to the new one.
• A visitor simply mistypes a URL directly into the address bar.

Do 404 errors actually hurt SEO?

Not by themselves, and not in the way many business owners fear. A normal, modest number of 404 errors is an expected, unavoidable part of running any website that's been live for more than a short while — pages come and go, other sites link to outdated URLs, people mistype addresses. Google has been explicit that encountering 404s during crawling is normal and isn't treated as a site-wide ranking penalty.

Where 404s do create real cost, it's indirect rather than a direct penalty:

• Lost traffic and link equity. If a page that still gets real visitor traffic, or has genuine backlinks pointing at it, returns a 404, you're losing both the visitors and the ranking value those backlinks were passing — value that a proper 301 redirect could have preserved.
• Worse user experience. A visitor who lands on a dead page with no clear next step is more likely to leave the site entirely rather than navigate to what they actually needed.
• Wasted crawl budget at scale. On a large site with a genuinely high number of dead URLs, crawlers spending time repeatedly requesting broken pages is time not spent discovering and re-crawling your genuinely important, live content. For most small business sites this isn't a significant concern, but it becomes more relevant as a site grows.

404 vs. "soft 404": a different, sneakier problem

A soft 404 is a page that visually looks like an error — it might say "Sorry, this page doesn't exist" or "No results found" — but the server actually returns a 200 (success) status code instead of a proper 404. From a search engine's perspective, this is confusing: the page claims to have succeeded while the content says there's nothing there. Search engines increasingly try to detect this pattern and may still choose not to index the page, or flag it in Search Console as a soft 404 issue, but relying on that detection rather than returning the correct status code in the first place is worse than just configuring the server to respond properly. Empty search-results pages, expired product listings, and misconfigured custom error pages are the most common sources of soft 404s.

How to find 404 errors on your site

• Check Google Search Console's Pages report (under the Indexing section) for URLs flagged as "Not found (404)" — this shows what Google itself has actually encountered while crawling.
• Run a site crawl with a crawling tool to find internal links pointing at dead pages across your entire site, not just ones Google has already found.
• Review server or analytics logs periodically for real 404 hits from actual visitors, which tells you which dead URLs are actually costing you traffic.

For a more detailed, step-by-step process, see our guide on finding and fixing broken links and 404 errors.

What to actually do about a 404 once you find one

The right response depends on whether there's a genuinely relevant replacement for the content that used to live there.

• If there's a relevant replacement page, set up a 301 redirect from the dead URL to it — see our guide on what a 301 redirect actually does and when to use one.
• If the content is genuinely gone with no equivalent, let the URL return a proper 404 (or a 410, if you want to explicitly signal it's permanently gone) rather than forcing an irrelevant redirect.
• Either way, make your 404 page itself useful — clear navigation, a search box, and links to your most important pages, so a visitor who lands on a dead URL isn't stranded. See our guide on custom 404 page best practices for what that should actually include.

The short version

A 404 error means a URL doesn't exist, and having some is completely normal — it's not a penalty and not something to chase down to zero. What matters is checking Search Console and your own site periodically, redirecting the ones that have real traffic or backlinks to a genuinely relevant page, and making sure your 404 page itself gives a lost visitor a clear way to find what they actually needed.