A 404 error is the response a web server sends when it has no page at the address a visitor or crawler asked for, and on its own it does not lower your rankings. It turns into an SEO problem when a page you still want found returns it, when a link or sitemap you control points at it, or when a missing page returns a normal 200 response and becomes a soft 404. This guide explains what each status means, how Google treats it and how to find, fix and redirect broken URLs.
See Aemorph's technical SEO service
Originally published 15 February 2024 on aemorph.com/seo/technical/404-page/. Updated October 2026.
What Is a 404 Error?
A 404 (Not Found) is the HTTP status code a server returns when it cannot find the resource at the requested URL. MDN Web Docs, which follows the HTTP standard RFC 9110, adds that links leading to a 404 page are often called broken or dead links (MDN, 404 Not Found). The page a visitor sees, often a "page not found" message, is a separate thing from the status code: a server sends the code in the response header, and the message is only the page body.
How is a 404 different from a 410?
A 410 (Gone) says the page is gone and the condition is likely to be permanent, while a 404 says only that nothing was found at that address. MDN advises a 404 when you do not know whether the removal is temporary or permanent (MDN, 410 Gone). For Google Search the choice makes no practical difference: Search Console Help states that Google currently treats a 410 the same as a 404.
Which other 4xx status codes matter for SEO?
Google's crawlers treat every 4xx status code except 429 the same way: the content does not exist, the URL leaves the index and crawling of it slows down. A 429 (Too Many Requests) is the exception, because Google reads it as a sign of an overloaded server and handles it like a server error.
| Status code | Name | What it tells the browser | Typical cause on a website | How Google's crawlers treat it |
|---|---|---|---|---|
| 400 | Bad Request | The server cannot process the request as sent | A malformed URL or request | Content treated as missing; URL removed from the index |
| 401 | Unauthorized | The resource needs valid login details | A page behind a login | Content treated as missing; URL removed from the index |
| 403 | Forbidden | The server refuses access to the resource | Folder permissions or a firewall rule | Content treated as missing; URL removed from the index |
| 404 | Not Found | Nothing exists at this URL | A deleted or moved page, or a mistyped link | Content treated as missing; URL removed from the index; crawled less often over time |
| 410 | Gone | The resource is gone, likely for good | A page removed on purpose | Treated the same as a 404 |
| 429 | Too Many Requests | The client sent too many requests | Rate limiting on the server or CDN | Treated as a server error; Google slows its crawling |
How common 4xx codes affect Google's crawlers, from Google's crawling documentation.
Sources: MDN Web Docs, October 2026 and others
The earlier version of this page described a 410 as a resource "moved to a different location". That describes a redirect, not a 410: a moved page answers with a 301 or 308 that sends the visitor on to the new address.
What Causes 404 Errors?
Most 404 errors come from a page that was deleted, moved or renamed without a redirect, from a link that points at a mistyped address, or from a server or CMS rule that sends requests to the wrong place.
- The page was deleted. A product, post or landing page came down, and nothing replaced it.
- The page moved. A redesign, a new URL structure or a site migration changed the address, and the old address has no redirect.
- A link is wrong. An internal link, a menu item or a link on another website points at an address with a typo, an old path or a missing character.
- A visitor typed the address. A misspelt URL in the browser bar returns a 404 like any other missing page.
- The server is misconfigured. A rewrite rule, a changed folder or a CMS setting sends requests for real pages to a path that does not exist.
Do URLs that never existed matter?
No: Google says 404s reported for URLs that never existed on your site are safe to ignore. Crawlers find odd addresses in scraped copies of your pages, in broken links on other sites and in malformed code, and a 404 for those addresses is the correct answer (Google Search Central Blog, Do 404 errors hurt my site?).
Do 404 Errors Hurt SEO?
404 errors on their own do not harm your site's indexing or ranking, according to Google's Search Console Help, and a 404 is the right response for a page you deleted with no replacement. The cost sits elsewhere: the page that returns the 404 leaves Google's index, visitors who reach it land on a dead end, and links from other sites that point at it no longer lead to a live page (Search Console Help, 404 errors).
| What happens | Effect on the missing URL | Effect on the rest of the site | What to do | Priority |
|---|---|---|---|---|
| A page you deleted returns 404 | Removed from Google's index over time | None; the rest of the site keeps its rankings | Leave the 404, or return a 410 | Low |
| A page you still want found returns 404 | Removed from the index, so it stops ranking | Visitors and links that pointed at it lose their destination | Restore the page or redirect it | High |
| An internal link or sitemap points at a 404 | Google keeps finding and requesting a dead address | Visitors following your own links hit a dead end | Correct the link or the sitemap entry | High |
| A moved page with links from other sites returns 404 | Removed from the index | Those links stop leading to a live page | 301 redirect to the new address | High |
| A URL that never existed returns 404 | Nothing to lose | None | Nothing | None |
What a 404 costs and what it does not, based on Google's Search Console Help.
Source: Google Search Console Help, October 2026
The earlier version of this page claimed that a 404 makes crawlers stop crawling the rest of a website and that bounce rate counts against you as a ranking signal. Google's crawling documentation says neither: it removes the missing URL and crawls that URL less often, and nothing in it penalises the rest of the site.
Which 404 errors need fixing?
Fix the 404s you created: pages you link to yourself, URLs listed in your sitemap, and moved pages that still attract visitors or links from other websites. Search Console Help recommends fixing only 404 errors you link to yourself or list in a sitemap, and redirecting a page that has moved. Aemorph adds one more group to that list: old URLs that other websites still link to, because Aemorph treats links to dead pages as carrying no value to the live site until a redirect reclaims them.
Does Google keep crawling a URL that returns 404?
Yes: Googlebot keeps requesting a 404 URL for some time and crawls it less and less often, and no setting tells Google to forget it permanently. The Search Console Page indexing report states this under "Not found (404)", and adds that a 404 is not necessarily a problem when the page was removed with no replacement (Search Console Help, Page indexing report). Aemorph's guide to how search engines work explains how Google crawls, renders, indexes and ranks pages.
What Is a Soft 404 Error?
A soft 404 is a page that tells visitors the content does not exist but answers with a 200 (success) status code, so search engines receive an error page dressed up as a working one. Google also counts a page with no main content, or an empty page, as a soft 404. Such pages are excluded from Search, and Search Console lists them under "Soft 404" in the Page indexing report (Google Search Central, Troubleshoot crawling errors).
What produces soft 404 errors?
Soft 404s come from a site or server that shows an error without sending an error code, or from a redirect that sends every missing page to the home page. Google lists a missing server-side include file, a broken database connection, an empty internal search results page and a JavaScript file that failed to load. Search Console Help adds a site that redirects unknown URLs to its home page instead of returning a 404. On large sites, soft 404 pages also keep being crawled and waste the crawl budget (Google, Crawl budget management).
How do you fix a soft 404?
Match the response to what happened to the content: a 404 or 410 when it is gone, a 301 when it moved, and a rendering fix when the page is real.
| Situation | Example | Right response | Status code to send | How to check it |
|---|---|---|---|---|
| The page and its content are gone with no replacement | A discontinued product with no successor | Return a not-found or gone response | 404 or 410 | Request the URL and read the status code in the response header |
| The page moved or has a clear replacement | A service page renamed in a redesign | Redirect to the new address | 301 | URL Inspection tool in Search Console |
| The page is real but Google flagged it as a soft 404 | A page whose content loads from a blocked script | Fix what stops it rendering: blocked or failed resources, too many resources, server errors, slow or very large files | 200, with the content visible | URL Inspection tool, View tested page, to see the rendered page and the status code |
Soft 404 fixes, from Google Search Central.
Source: Google Search Central, October 2026
How do single-page apps avoid soft 404s?
In a client-side rendered single-page app the browser handles routing, so a meaningful status code for a missing page is often impossible or impractical, and Google recommends either a JavaScript redirect to a URL that returns a real 404, such as /not-found, or a robots noindex tag added to error views. Both options appear in Google's JavaScript SEO basics, with sample code for each (Google Search Central, JavaScript SEO basics).
How Do You Find 404 Errors on Your Site?
Combine Google Search Console for the URLs Google found, a site crawler for the broken links on your own pages, a backlink tool for the dead URLs other websites still link to, and your server logs for everything else.
- Search Console Page indexing report. Open "Not found (404)" and "Soft 404" to see the URLs Google tried and failed to index, with examples for each.
- URL Inspection tool. Check a single URL: the status code Google received, how Google rendered it and where Google found it.
- A site crawler. A crawler such as Screaming Frog reads your site the way a search engine does and lists every link that returns a 4xx code, with the page that links to it.
- A backlink tool. Ahrefs and similar tools report links from other websites that point at URLs returning 404, so you know which dead pages still carry links.
- Server logs. Access logs show every request that returned 404, from visitors and crawlers alike, including addresses no report shows.
Bing's guidelines ask you to return a 404 for removed pages and to report deleted URLs through IndexNow (Bing Webmaster Guidelines).
How Do You Fix a 404 Error?
Decide for each URL whether the content moved, was replaced or is gone: redirect it with a 301 when a close equivalent exists, correct the link when your own link is wrong, and leave a true 404 or 410 when nothing replaces it. Search Console Help also recommends mapping common misspellings of a real URL to that page with a redirect.
| Situation | Example | Fix | Status code | Why |
|---|---|---|---|---|
| The page moved to a new address | /services/seo-audit.html became /services/seo-audit/ | Redirect the old URL to the new one | 301 | Visitors and Google reach the new page; Google uses a 301 as a strong signal that the target is the one to show |
| The page was replaced by a close equivalent | An old offer page replaced by a current one | Redirect to the replacement | 301 | The old address keeps serving the visitor's need |
| Your own link or menu points at a typo | A menu link to /contcat/ | Correct the link at its source | None after the fix | A redirect only patches a link you control |
| Your sitemap lists a dead URL | A deleted post still in sitemap.xml | Remove it from the sitemap | None after the fix | A sitemap lists only live pages you want indexed |
| A common misspelling of a real URL | /cancelling/ typed as /canceling/ | Redirect the misspelling to the real page | 301 | Visitors who mistype still arrive |
| The content is gone and nothing replaces it | A one-off event page | Leave the 404, or return a 410 | 404 or 410 | The correct answer for a removed page; redirecting it to the home page creates a soft 404 |
Which fix fits which 404.
Which redirect fits each move?
Use a server-side 301 or 308 for a permanent move and a 302 or 307 only for a short, temporary one. Google treats 301 and 308 as permanent redirects that show the new address in search results, and 302, 303 and 307 as temporary redirects that keep the old address in results (Google Search Central, Redirects and Google Search). Bing's guidelines give the same split and reserve 302 redirects for "very short-term changes (less than 2 days)" (Bing Webmaster Guidelines).
Why do redirect chains matter?
Long redirect chains have a negative effect on crawling, according to Google's crawl budget guide, and Google's crawlers follow up to 10 redirect hops by default. Every hop is one more request before Google reaches the page you want indexed. Point every old URL straight at its final destination, and update old redirects when a page moves a second time.
Sources: Google Search Central, October 2026 and others
What Makes a Good Custom 404 Page?
A good custom 404 page tells the visitor plainly that the page is missing, keeps your site's look and navigation, and offers clear next steps, while the server still returns a 404 status code. Google's own tips for a useful 404 page cover four points (Google Search Central, Troubleshoot crawling errors):
- Say clearly that the page cannot be found, in friendly language.
- Keep the same look and navigation as the rest of your site.
- Link to your most popular pages and to the home page.
- Offer a way to report the broken link.
What else helps visitors on a 404 page?
A site search box, links to your main categories or best-selling products, and a contact link turn a dead end into a route back into the site. These additions come from the earlier version of this page and remain sound practice: a visitor looking for a missing product page finds a related one, and a visitor looking for help reaches your team. A touch of brand humour fits some brands, as long as the message stays clear.
Does the custom 404 page need a 404 status code?
Yes: Google states that custom 404 pages exist only for visitors and are useless to search engines, so the server returns a 404 status code to keep them out of the index. A styled error page that returns 200 is the soft 404 described above. Check the status code in the response header after every change to your CMS, theme or server configuration, since the page a visitor sees does not show which code the server sent.
Keep Broken URLs From Costing You Visitors
404 errors stay harmless when every page you still want found returns 200, every moved page redirects in one hop, and every removed page returns a real 404 or 410. Check the Search Console Page indexing report each month, crawl the site after every redesign or migration, and reclaim dead URLs that still carry links with one-to-one redirects. If you prefer an experienced team to run this for you, Aemorph's technical SEO service crawls your site with Screaming Frog, lists status code faults, redirect chains and broken links, and specifies each fix at the template for your developers, who set the pace. The SEO audit covers the same checks inside a wider review, and the on-page SEO audit guide shows the page-level checks that sit beside them. Aemorph's guide to web design and development sets out how to plan, build and launch a site with search built in.