Updated
To fix 404 errors in Google Search Console, decide for each URL whether it should exist. If the page moved, add a 301 redirect from the old URL to the new one. If it was removed on purpose and nothing replaces it, leave the 404 in place and remove any links to it from your own pages and sitemaps. Google says 404s don’t count against your site, so only the ones you link to yourself, list in a sitemap, or that used to get visits or backlinks are worth your time.
What “Not found (404)” means in Search Console
“Not found (404)” is a reason under “Why pages aren’t indexed” in the Page indexing report (Indexing > Pages). Google requested the URL and your server answered with a 404 status, meaning the page doesn’t exist. Pages that return a 404 aren’t indexed, and URLs already in the index are removed once Google sees the 404.
Google’s own help page is clear that 404 responses are not necessarily a problem if the page was removed without a replacement, and recommends fixing only the 404s that you link to yourself or list in a sitemap. If the page moved, it recommends a redirect.
Do 404 errors hurt your rankings?
Not by themselves. Google’s Search Central team wrote that 404s are a normal part of the web and don’t count against you in search results. Google’s crawling documentation adds that 4xx errors (other than 429) have no effect on crawl rate. Missing pages don’t drag down the pages that still exist.
What 404s can cost you is indirect: visitors who hit a dead end from your own links, and the value of backlinks pointing at a page that’s gone. That is why the triage below is about where each 404 came from, not how many there are.
Which 404s to ignore and which to fix
| Situation | What to do |
|---|---|
| URL never existed (a typo in someone else’s link, a scraper, a hacked-site leftover) | Ignore it. The 404 is correct. |
| You deleted the page, it had no visits or backlinks, and nothing replaces it | Leave the 404 (or 410). Remove it from your sitemap and internal links. |
| The page moved or was renamed | 301 redirect the old URL to the new one, in one hop. The old URL then shows as Page with redirect, which is fine. |
| You deleted the page but it had traffic or backlinks, and a close equivalent exists | 301 redirect to that equivalent page. |
| Your own internal link or sitemap points at it | Fix the link or the sitemap entry. This is the 404 Google wants you to fix. |
| The page should exist but returns 404 | Fix the page: broken permalinks, a routing rule, a plugin conflict, a listing set to draft by mistake. |
| Hundreds of junk URLs with odd parameters or paths | Usually spam links or an old hack. A 404 or 410 is right; don’t redirect them. |
Don’t redirect every 404 to your home page. Google’s John Mueller has said Google mostly treats those redirects as 404s anyway, and they confuse visitors. If there’s no real replacement, a helpful 404 page with search and links to your main sections is better.
How to find the page linking to a 404
Google finds 404 URLs through links: on your site, in your sitemaps, or on other sites. To find out which:
- Inspect the URL. Click a URL in the Not found (404) list, then Inspect URL. Expand Page indexing and look under Discovery: Sitemaps lists any submitted sitemap that contains it, and Referring page shows a page Google may have found it on. Google says the referring page is one it possibly used, so it can be old or missing (URL Inspection help).
- Filter the report to your sitemaps. Switch the Page indexing report from “All known pages” to “All submitted pages”. A 404 that appears there is in your sitemap, and the sitemap needs regenerating.
- Crawl your own site with a site crawler to list internal links that return 404. Fix them in the template, not page by page.
- Check Google Analytics. Visits to your 404 page, broken down by page referrer, show the links people actually click. A 404 that gets real visits is worth a redirect even if Google never mentioned it.
- Check backlinks. If other sites link to the dead URL, a redirect to the closest live page keeps those visitors and that link.
How to fix 404 errors, step by step
- Export the list. Open Indexing > Pages, click Not found (404) and export. The report shows up to 1,000 example URLs per reason.
- Group by pattern. Sort the URLs by path. Forty URLs under
/old-blog/are one decision, not forty. - Decide per group using the table above: leave, redirect, fix the link, or restore the page.
- Add redirects at the server or CMS level, as permanent (301) redirects straight to the final URL. Avoid chains from earlier migrations.
- Fix your own links and sitemaps. Regenerate sitemaps so they contain only live, indexable URLs.
- Click Validate fix on the Not found (404) issue once the URLs you care about are fixed. Google rechecks them; it says validation typically takes up to about two weeks and sometimes longer.
- Expect some to stay. Google may keep recrawling an old URL for a long time because it still knows about it. A 404 that Google keeps listing, for a page you meant to delete, is not a problem to solve.
404 vs 410, soft 404 and blocked pages
| Status | What it means | What Google does |
|---|---|---|
| 404 Not found | The page doesn’t exist | Drops it from the index; may recheck the URL later |
| 410 Gone | The page existed and was removed on purpose | Google treats 410 the same as 404; use whichever your platform supports |
| Soft 404 | The page says “not found” or is empty but returns 200 | Reported as Soft 404 and not indexed. See soft 404s. |
| 403 / 401 | Access forbidden or login required | Reported as Blocked due to access forbidden (403) or unauthorized request (401) |
| Blocked by robots.txt | Google isn’t allowed to crawl the URL | Reported as URL blocked by robots.txt. See blocked by robots.txt. |
A 404 doesn’t mean Google or anyone else blocked you. It is your server saying the page isn’t there. If you can open the page in a browser but Google reports a 404, check whether the server answers bots differently, whether the URL in the report has a slightly different spelling or trailing slash, and run Test live URL in URL Inspection to see what Google gets today.
Removed listings on directory websites
On a directory most 404s come from listings and the pages built from them. Some decisions to make once, as rules, instead of URL by URL:
- Duplicate listings merged: 301 the duplicate to the listing that stays.
- Listing slug changed (
/listing/joes-plumbing-llcto/listing/joes-plumbing): 301 the old slug. Many platforms do this automatically; check yours does. - Business closed, no replacement: a 404 or 410 is correct. If the old listing still gets searches, an alternative is to keep the page live with a clear “this business has closed” notice and real links to similar businesses nearby. Don’t keep an empty shell, which Google will likely report as a soft 404.
- Expired paid listings: avoid flipping a page between 200 and 404 each time a member lapses and renews. Keep a free version of the listing live, or decide that lapsed listings are removed for good.
- Empty category-in-city pages (
/plumbers/smalltown-txwith no plumbers): a 404 for combinations you never meant to publish is fine. Just don’t link to them or list them in the sitemap. - Pagination past the last page (
/plumbers/austin-tx?page=40when there are 12 pages): a 404 is fine.
After a restructure, a sudden jump in Not found (404) almost always means old URLs weren’t mapped. That’s a migration problem, covered in the website migration checklist.
How DirectoryLens helps
- Visitor actions lists pages showing “not found” that get visits, from Google Analytics, with the page each visit came from. Those are the broken links people actually click, and the ones worth a redirect first. It finds them by page title, so your 404 page’s title needs to contain “404” or “not found”.
- URL changes checks the old URLs after a URL change and flags any that end in a 404 or other error page, with the visits each had before.
- Google indexing > Individual pages can inspect every URL in a sitemap through Google’s URL Inspection API. Any that Google reports as Not found (404) show up under Other not indexed with Google’s coverage state, so you can see which sitemap entries are dead.
- Bing, if you link Bing Webmaster Tools, shows Bingbot’s daily 4xx and 5xx counts next to your Google data.
For every other status in the Page indexing report, see the page indexing issues guide.