404 SEO

A 404 status code means the server cannot find the resource at the requested URL. That is all. It does not automatically hurt your site’s SEO. Google’s documentation and John Mueller have repeatedly said that most 404 errors are not worth fixing and do not harm indexing or ranking. The trap is treating every 404 as a crisis. I have seen SEOs redirect all 404s to the homepage, or block them with robots.txt, or create thin content to replace them. All of those are worse than leaving the 404. The only 404s that matter are the ones on important URLs: pages with backlinks, pages listed in your sitemap, pages that get significant traffic, or pages that are linked from important internal pages. For those, you have three options: restore the content, redirect to a relevant replacement, or explicitly signal that it is gone with a 410 status. Everything else can stay as a 404. My rule: check Google Search Console for the 404s that appear in the Index Coverage report, ignore the rest. Use an [SEO audit](/seo-audit/) to find them, but do not automate fixes. For a client site with 5,000 404s, I found that only 47 had backlinks; the rest were harmless.

The decision to fix a 404 depends on why the URL exists. If the URL is in your sitemap, Google expects it to be accessible. A 404 there tells Google the page is not available, which can waste crawl budget and delay indexing of other pages. Similarly, if the URL has acquired backlinks, those links are now pointing to a dead end. A [301 redirect](/301-redirect/) to the most relevant replacement passes the link equity. If you have moved the content to a new URL, use a 301 redirect. If the content is permanently removed with no close replacement, use a 410 instead of a 404. Google treats both similarly, but a 410 may speed up removal from the index. Be careful not to redirect all 404s to the homepage; that confuses users and is seen as a soft 404 by Google. Also, never block 404s with [robots.txt](/robots-txt/) — that prevents Google from seeing the 404 and may cause the URL to be treated as a soft 404. A custom 404 page with navigation and a search box can help users who land on a dead end. For sitemap management, see [sitemap](/sitemap/). I once worked on a site migration where we forgot to redirect a product page that had affiliate links from three major review sites. The 404 dropped its organic traffic by 80% in two weeks.

For years, I treated every 404 as a problem. I would redirect them all to the homepage, thinking I was preserving link equity. I was wrong. Google’s guidelines explicitly advise against redirecting all 404s to the homepage because it creates a poor user experience and can be seen as a soft 404. I also used to block 404s from crawling using robots.txt, which is a mistake. Blocking them prevents Google from recognising the 404 and may cause the URL to be indexed as a soft 404 page. I have since realised that most 404s are harmless and should be left alone. The real issue is the ones that matter. I also did not know the difference between 404 and 410. I used 404 for everything. Now I use 410 when I have permanently removed content and do not want it to be re-crawled as a 404. Another mistake: I did not check the Index Coverage report in Search Console. I was looking at the Crawl Errors report instead. The Index Coverage report is more accurate because it shows which 404s are on pages that Google tried to index. For more on managing indexation, see [canonical tag](/canonical-tag/) and robots.txt. I once redirected a site's 404s to the homepage and saw a 15% drop in traffic from those pages.