Every website collects 404 errors over time. Products sell out, blog posts are merged, campaigns end and someone mistypes a link in a newsletter. Most of those 404s are harmless. A handful, though, quietly cost you traffic, backlinks and sales, and they are easy to miss without the right process.
This guide shows you how to separate harmless 404s from harmful ones, how to find broken pages using free and paid tools, and exactly how to fix each type. You will also learn how to spot soft 404s, how to build a helpful custom 404 page, and which mistakes to avoid when cleaning up.
Key Takeaways
- A 404 status is the correct response for a page that does not exist; not every 404 needs fixing.
- Fix 404s that receive internal links, backlinks or real traffic, usually with a 301 to a relevant page.
- Soft 404s (error pages returning 200 OK) confuse Google and should be corrected.
- Never mass-redirect 404s to the homepage.
- Use Search Console, a site crawler, server logs and a backlink tool together for full coverage.
What Is a 404 Error?
A 404 is an HTTP status code meaning “Not Found”. The server is reachable, but there is nothing at the requested URL. Its sibling, 410 “Gone”, says the page was removed intentionally.
“All 4xx errors, except 429, are treated the same: Google crawlers inform the next processing system that the content doesn't exist.”
— Google Search Central, How HTTP status codes and network errors affect Google Search
According to the same documentation, URLs that already were indexed and now return a 4xx are removed from the index, newly found 404 URLs are not processed, and crawl frequency for them gradually decreases. In other words, Google treats a 404 as information, not as a punishment.
Do 404 Errors Hurt SEO?
A 404 on a page that should not exist does not damage the rest of your site. The real costs come from specific situations:
- Lost link value: if external sites link to a URL that now 404s, the signals from those links no longer reach a live page.
- Broken user journeys: visitors who hit a dead end often leave rather than search for what they wanted.
- Wasted crawling on large sites: thousands of broken internal links send crawlers to dead URLs. Google's crawl budget guide recommends returning a 404 or 410 for permanently removed pages so that crawlers do not keep revisiting them as live content.
- Accidental removals: a deployment bug that turns your best category page into a 404 is a ranking problem, not a housekeeping one.
404 vs 410 vs Soft 404 vs Redirect
| Response | What it tells Google | When to use it |
|---|---|---|
| 404 Not Found | Content does not exist at this URL | Removed pages with no replacement; mistyped URLs |
| 410 Gone | Content was removed intentionally (treated like other 4xx) | When you want to state removal is permanent and deliberate |
| Soft 404 (200 OK on an error or empty page) | Mixed signal: “page exists” but content says it doesn't | Never; fix it |
| 301 redirect | Content moved permanently to a new URL | Moved or merged pages with a clear equivalent |
| 200 OK with real content | Page is live | Restored pages that were removed by mistake |
What Is a Soft 404?
Google's troubleshooting documentation describes a soft 404 as a URL that returns a page telling the user it does not exist, while still responding with a 200 OK status. Empty category pages, “no products found” templates and thin error messages are common causes. JavaScript apps are particularly prone to them because the server returns 200 for every route and the app decides later that the content is missing; see our JavaScript SEO guide for the fixes.
Google's recommended fixes depend on the situation:
- If the page and content are gone with no replacement, return a 404 or 410.
- If the page has moved or has a clear replacement, use a 301 redirect.
- If the page and content still exist, the page probably did not load properly for Googlebot, so check rendering with the URL Inspection tool.
How to Find 404 Errors
| Tool | What it finds | Cost |
|---|---|---|
| Google Search Console (Page indexing report) | URLs Google tried to crawl that returned 404 or soft 404 | Free |
| Site crawler (Screaming Frog, Sitebulb, Semrush Site Audit) | Broken internal links and the pages linking to them | Free tier / paid |
| Server or CDN logs | Real requests from users and bots hitting 404s | Usually free |
| Backlink tool (Ahrefs, Semrush, Moz) | External links pointing at dead pages | Paid |
| GA4 (page title of 404 template) | Which broken URLs real visitors land on | Free |
In Search Console, open Indexing → Pages and check the “Not found (404)” and “Soft 404” rows. Our walkthrough of Google Search Console explains how to navigate these reports. In GA4, make sure your 404 template has a distinctive page title (for example “Page not found”) so you can report on it; our GA4 beginner's guide shows how to build that exploration.
How to Fix 404 Errors: Step by Step
- Export all 404 URLs from Search Console, your crawler and your logs into one sheet and remove duplicates.
- Add context for each URL: number of internal links, number of referring domains, and recent visits.
- Decide intent. Was the page removed on purpose, moved, or broken by mistake?
- Restore pages deleted by mistake so they return 200 again.
- 301 redirect moved or merged pages, and removed pages with backlinks, to the most relevant live page. Our guide on 301 vs 302 redirects covers implementation.
- Fix internal links at the source so they point directly to the correct URL instead of relying on a redirect.
- Let the rest return 404 or 410. Old junk URLs, spam URLs and typos with no links need no action.
- Update your XML sitemap so it lists no dead URLs; see our XML sitemap guide.
- Validate the fix in Search Console and recrawl to confirm.
Quick check from the command line
curl -I https://www.example.com/old-page/
HTTP/2 404If a page that should be missing returns HTTP/2 200, you have a soft 404 to fix.
Reclaiming Backlinks From Dead Pages
Pages that once earned links are the most valuable 404s to fix. Sort your list by referring domains, then redirect each URL to its closest current equivalent. If no equivalent exists, consider recreating an updated version of the content. The same research works in reverse as an outreach tactic, which we cover in our guide to broken link building.
Build a Helpful Custom 404 Page
Some 404s will always reach users, so make the page useful. Google notes that custom 404 pages are created for users; just make sure the template still returns a real 404 status.
- Say clearly that the page could not be found, in friendly language.
- Keep your normal header, navigation and footer.
- Add a search box and links to popular pages or categories.
- Offer a contact option for visitors who are stuck.
- Keep it lightweight and fast.
Common Mistakes When Fixing 404s
- Redirecting every 404 to the homepage. It frustrates users and can be treated as a soft 404, so it rarely passes any benefit.
- Returning 200 on the error template. This creates soft 404s at scale. Check the status code, not just the design.
- Blocking 404 URLs in robots.txt. If Google cannot crawl them, it cannot see they are gone. Our article on robots.txt vs meta robots explains why blocking and de-indexing are different jobs.
- Chasing every 404 in Search Console. Many are junk URLs from scrapers or old typos. Prioritise by links and traffic.
- Removing redirects too soon. Old URLs start 404ing again and the lost value returns.
Related Guides
- Duplicate Content in SEO: Causes and Fixes
- Hreflang Tags: A Guide to International SEO
- JavaScript SEO: How to Make JS Sites Search-Friendly
Frequently Asked Questions
Do 404 errors hurt SEO?
A 404 on a page that genuinely no longer exists is normal and expected. Problems arise when important pages 404 by mistake, when internal links point to dead URLs, or when pages with valuable backlinks disappear without a redirect.
Should I redirect every 404 to my homepage?
No. Blanket homepage redirects confuse users and Google may treat them as soft 404s. Redirect only to a genuinely relevant replacement; otherwise let the URL return 404 or 410.
What is the difference between 404 and 410?
A 404 means not found; a 410 means gone on purpose. Google's documentation groups all 4xx codes (except 429) together, so both remove the URL from the index over time. Use 410 if you want to be explicit that removal is intentional.
What is a soft 404?
A soft 404 is a page that tells users the content does not exist (or is essentially empty) but returns a 200 OK status code. Google flags these in Search Console and they should return a real 404/410 or redirect to a relevant page.
How do I find 404 errors on my site?
Use Search Console's Page indexing report (Not found 404 and Soft 404), a site crawler for broken internal links, server logs for real hits, and a backlink tool for external links pointing at dead pages.
How long does it take Google to drop a 404 page?
There is no fixed timeline. Google gradually crawls 404 URLs less often and removes them from the index once it has confirmed they are gone.
Conclusion
404 errors are a normal part of a living website. The job is not to eliminate them all, but to make sure no important page, valuable backlink or common user path ends in a dead end. Review your 404 data monthly, fix internal links at the source, redirect only where there is a genuine match, and make your custom 404 page helpful. A broken-page review should also be a standard step in every SEO audit.
Need help cleaning up broken pages after a migration or redesign? Our on-page optimization and technical SEO teams can audit and fix them for you. Get a free quote today.
References
- Google Search Central: How HTTP status codes and network errors affect Google Search
- Google Search Central: Troubleshoot Google Search crawling errors (soft 404s)
- Google Search Central: Redirects and Google Search
- Google Search Central: Understand JavaScript SEO basics
- Google Crawling Infrastructure: Large site owner's guide to managing your crawl budget
- Semrush: What Is a 404 Error? How It Affects SEO and How to Fix It
