Brendan Tully0 Comments
October 7, 2026
A long list under “Not found (404)” in Google Search Console looks alarming, and plenty of site owners respond by redirecting every URL in it. That usually makes things worse. A 404 is the correct answer for a page that does not exist, and Google expects to see them. The SEO work is in separating the harmless 404s from the ones that waste crawling, leak link value or point to a fault in the site itself.
The video walks through several real Search Console accounts: an old site with dead subdomains still being crawled, a site with junk URLs that look like a negative SEO attack, and a large Magento store with hundreds of retired product URLs. Below is what each pattern means for search.
Google does not mark a site down for having 404s. When a URL is gone, a 404 tells Google to stop spending time on it and drop it from the index. What counts is the status code, not the message on screen: the server has to send a 404 code to the browser, not a 200 OK with a page that happens to say “not found”. Some content management systems get this wrong, showing a friendly error page while returning 200, so Google sees what looks like a real but empty page.
The other trap is a plugin that redirects every 404 to the homepage or another page. The report goes quiet, but you are now telling Google that thousands of URLs which should not exist are valid. Over time that is bad for SEO, and it opens the door to negative SEO, because anyone can invent junk URLs on your domain and every one of them resolves. Let missing pages return a 404.
Work in the domain property rather than a URL-prefix property. The domain property covers http and https, with and without www, and every subdomain, so nothing is hidden.
On older sites the host list in Crawl stats often includes subdomains retired years ago that still have DNS records. In one account in the video, an old CRM subdomain returned 404 for 52 percent of the requests Google made to it. Redirects are not the answer here. Delete the DNS records for hosts that no longer exist so Google stops crawling them.
ads.txt and favicon.ico turn up constantly. ads.txt stands for authorised digital sellers, and if you do not sell advertising you do not strictly need it, but a blank file stops the repeated 404. An empty file also states that nobody is authorised to sell ad inventory on your site. A missing favicon is a five-minute fix. Requests under the .well-known folder for mobile app links can safely 404 if you have no app.
This is where ranking value sits. E-commerce sites add and retire products constantly, and old product URLs keep the links they picked up over the years. If a product has been replaced, 301 redirect its URL to the replacement or the closest category, so the link value and the visitor both land somewhere useful. The same applies to old landing pages, PDFs and pages from previous versions of the site. A URL that Google crawls often usually has links pointing to it from somewhere.
Large numbers of nonsense URLs, especially from a spammy referring site, can be a negative SEO attack. These should keep returning 404. Click into individual URLs to see where Google found them, then open the Links report in Search Console and review the top linking sites. If a domain is clearly pointing garbage at you, consider disavowing it.
A broken link in a template or an internal link to a URL that does not exist is a site fault rather than a crawl curiosity. One example in the video traced back to an internal link pointing at a theme folder. Fix the link at its source.
On a big site, export the report to Google Sheets and filter it. Most URLs fall into one of four groups: genuine old pages that deserve a redirect, technical files to create, junk to leave as 404, and areas Google should not be crawling at all. For that last group a robots.txt Disallow rule is better than redirects. On the Magento store in the video, review list URLs for old products were a good candidate for blocking.
Do not chase a clean report for its own sake. On one tidy site in the video, the remaining 404s were natural ones, and the right call was to leave them alone.
Missing files in the template affect conversion as well as crawling. A browser that hits a 404 for a script mid-render can stall, and the page feels like it hangs halfway through. On WordPress, each 404 is processed by PHP, so a heavy flow of them adds server load at exactly the moment an email send or campaign drives traffic to the site.
After the changes, watch the share of 404 responses in Crawl stats, which should fall as Google recrawls. The count in the Pages report will drop more slowly, and URLs that correctly return 404 can sit there for a long time without doing harm. For redirected products and landing pages, track whether the destination pages pick up the traffic and links over the following months.
Comments are closed.