404 vs 410: How to Handle Removed Pages the Right Way

What 404 and 410 mean

Both 404 and 410 are HTTP status codes in the 4xx "client error" family. They describe the same broad situation, a request for a resource the server will not return, but they differ in certainty.

404 Not Found means the server cannot find a resource matching the request. It is deliberately open-ended. The page may never have existed, may have been moved without a redirect, may be temporarily unavailable, or may have been deleted. The 404 makes no promise about whether the URL will return.

410 Gone is more specific. It means the resource was available before and has been intentionally removed, with no forwarding address, and the condition is expected to be permanent. In plain terms, 404 says "I cannot find this," while 410 says "this used to be here and it is gone for good."

For human visitors the two are nearly identical, since most servers render the same error page for both. The distinction matters most to crawlers and to the engineers deciding how to retire a URL.

How Google treats each

Google has been consistent on this for years. Both 404 and 410 lead to the same outcome: the URL is eventually removed from the index. A 410 is treated as a slightly more definitive signal, and Google may stop attempting to recrawl it a little sooner than a 404, which it sometimes revisits in case the page comes back. But the gap is measured in a small number of recrawl cycles, not in ranking impact.

The important point is that neither code carries a penalty. Removing a page that genuinely has no equivalent is normal site maintenance, and search engines expect it. Google does not lower the rankings of your remaining pages because some URLs return 404 or 410. Treat the choice between the two as a question of how quickly and clearly you want the URL flushed from the index, not as a ranking lever.

One caveat: Googlebot can be slow. Even a clean 410 may linger in Search Console reports for a while before it clears, and a cached snippet can persist briefly. That is expected behavior, not a problem to fix.

404, 410, or 301: how to choose

The decision rests on one question: does an equivalent page exist that satisfies the same user intent?

A useful rule: redirect for continuity, return a gone status for closure. Redirecting an irrelevant URL to a loosely related page, or to the homepage, creates a poor experience and is often interpreted as a soft 404 anyway, which defeats the purpose. If you would not send a human there expecting them to find what they wanted, do not redirect there. For the difference between permanent and temporary redirects, see our guide to 301 vs 302 redirects.

Implementation

Returning a 404 usually requires no work, since servers and content management systems emit it automatically for missing URLs. Returning a 410 is a deliberate act. Here is how to send a 410 for one URL or a small set of URLs.

Apache, in your .htaccess file or virtual host config:

# Return 410 Gone for a single removed URL
Redirect 410 /old-product-page/

# Return 410 for several removed URLs
<Location "/discontinued/widget-a/">
    Redirect 410
</Location>

# Match a pattern of removed URLs with RewriteRule
RewriteEngine On
RewriteRule ^expired-promo/ - [G]

The [G] flag in a RewriteRule is shorthand for "Gone" and emits a 410.

Nginx, in your server block:

# Return 410 Gone for a single removed URL
location = /old-product-page/ {
    return 410;
}

# Return 410 for a directory of removed content
location ^~ /discontinued/ {
    return 410;
}

# Optional: serve a friendly body with the 410
error_page 410 /gone.html;
location = /gone.html {
    internal;
}

Reload the server after editing config (nginx -s reload or apachectl graceful) and confirm the response with a header check, for example curl -I https://example.com/old-product-page/, which should show HTTP/1.1 410 Gone.

Handling removed pages at scale

When you retire many URLs at once, the status code is only part of the job. Good housekeeping around it is what keeps the site clean and the user experience intact.

Build a custom 404 page. The default server error page is a dead end. A branded 404 that returns the correct status code, offers a search box, and links to your main sections turns a mistake into a recovery path. Make sure it actually returns 404 and not 200, which would create a soft 404.

Drive removals from server config, not page logic. For large sets, manage 410 and 301 rules in the server or a centralized redirect map rather than scattering them across templates. This keeps the rules auditable and avoids conflicting handlers.

Remove internal links to dead URLs. Crawlers find dead pages through your own links. After you retire URLs, sweep the site for links pointing at them in navigation, body content, and footers, and either repoint or delete those links. Leaving them in place wastes crawl budget and frustrates visitors.

Keep your XML sitemap clean. A sitemap should list only live, indexable URLs that return 200. Remove retired URLs from it promptly. Listing a 404 or 410 in a sitemap sends mixed signals and can slow down how fast the removal is recognized. If you control crawler access more broadly, our robots.txt reference explains what belongs there and what does not.

Common mistakes

Redirecting everything to the homepage. This is the most common error when sites delete pages in bulk. A blanket redirect to the root sends users somewhere that does not answer their request, and Google generally treats these mass homepage redirects as soft 404s, so they do not preserve value and do not even keep the URL out of error reports. Redirect to a true equivalent or return a gone status. Do not split the difference with the homepage.

Soft 404s. A soft 404 is a page that looks like an error to a human but returns a 200 OK status to crawlers, or a thin "not found" message served with a success code. Google flags these because the signal contradicts the content. Empty category pages, expired listings that still render a stub, and JavaScript apps that show "no results" without setting the status are typical culprits. The fix is to return the honest code: 404, 410, or a 301 to a real replacement. We cover the diagnosis and repair in detail in our guide to soft 404 errors.

Deleting valuable pages without a plan. The myth that 404s hurt rankings persists because people notice traffic drops after deleting pages and blame the status code. The status code is not the cause. The loss comes from removing a page that earned links and traffic without redirecting its equity to a worthy successor. Before you delete, check whether the URL has backlinks or meaningful traffic. If it does, redirect it to the best equivalent rather than letting it go dark.

FAQ