404 Link Checker
Links that lead to a page that is not there. What separates a 404 from a 410 and a soft 404, how to find each one, and why a list of 404s is never the full list of broken links.
The 404 checker is not open yet
TagPipes is finishing continuous link monitoring first, and the one-off checker follows it. Rather than put a scan box here that cannot scan, this page is the part that is useful now. There is a form below if you want telling when it opens.
The free check that is open today is the Website Accessibility Checker: paste a URL, get a report in a few minutes, no signup.
What a 404 is, and what it is not
A 404 is a server saying: I exist, I understood your request, and there is nothing at that address. It is not an error in the sense of something being broken. It is the correct, honest answer to a request for something that is not there, and a site that returns 404 for a missing page is behaving properly.
That is worth holding onto, because the instinct when a report shows four hundred 404s is that four hundred things are wrong with the site. Usually what is wrong is much smaller: a handful of pages were removed or renamed, and every link that pointed at them now produces a row in the report.
The real question a 404 report answers is not how many there are. It is: which addresses are people and search engines still asking for, and should any of them be leading somewhere?
404, 410 and the soft 404
Three ways a site can respond to a request for a missing page, with very different consequences.
| Response | What the server is saying | What a crawler does | When to use it |
|---|---|---|---|
| 404 Not Found | Nothing here. Possibly temporary, possibly a typo, I am not committing either way | Comes back occasionally to check, for a long time | The default, and correct for almost everything |
| 410 Gone | This was here, it has been deliberately removed, and it is not coming back | Drops it faster and stops asking | Content you removed on purpose and will never restore, such as an expired offer |
| 301 redirect | It moved. Here is the new address | Follows it and transfers the link value | Anything renamed, migrated or consolidated. The most useful of the four |
| Soft 404 | 200 OK, here is your page, which is actually a not-found page | Believes you. Indexes a not-found page as real content | Never. This is always a bug, and it is the one that hides from your own reports |
Soft 404s are the ones worth hunting. Every tool you own believes the page is fine, including your own link checker, because the server said 200 and a checker has no way to know that the page it received is an apology.
Broken links that are not 404s
If you only look for 404s, these never appear on any list.
Soft 404s
The destination is gone and the server says 200. Invisible to every status-code check, and genuinely common on sites that built a friendly not-found page and wired it up wrong.
Timeouts and dead hosts
No status code at all, because nothing answered. Common on outbound links to sites that have quietly ended.
403 and 401
The server refused. Sometimes it refuses everyone, sometimes only robots, which is why a flagged 403 always needs a human to open it.
Redirect chains and loops
The link eventually works, through four hops, losing query parameters on the way. Nothing reports as broken and the tracking arrives empty.
Expired domains
Perfect 200 response, completely different site. Your link now works and recommends something you have never seen.
Right code, wrong page
A link that points at the wrong existing page. No tool will ever flag this, because from the outside nothing at all is wrong.
What to do with the 404s you find
Not every missing page deserves a redirect, and blanket redirects cause their own problems.
1. Redirect the ones with an obvious successor
A renamed page, a consolidated article, a product that was replaced. A 301 to the genuinely equivalent page keeps both the visitor and the link value.
2. Let the rest 404 properly
Redirecting everything to the homepage is worse than a 404. The visitor gets no explanation and search engines treat it as a soft 404 anyway. A good 404 page with search and a route back is the right answer for content that has no successor.
3. Use 410 sparingly and deliberately
For content you removed on purpose and will never bring back. It tells crawlers to stop asking, which is helpful and is also irreversible in practice, so it is not the default.
4. Fix the links you control
A redirect covers the links you cannot edit. For the links on your own pages, updating them is better: one fewer hop, no dependence on a redirect rule surviving the next migration.
Frequently asked questions
Tell me when the checker opens
One email when the free checker and the monitor are open to use. Nothing else, and no newsletter.
More on broken links
The same subject from the angle you came with, and the recurring version of the check.
- Broken Link CheckerWhat a link check does, what it reports, and what to fix first.
- Website Broken Link CheckerChecking a whole site: where broken links come from and how often to look.
- Dead Link CheckerLink rot: expired domains, dead hosts and pages that quietly stopped existing.
- Broken Link MonitorThe recurring version: scan, alert on new breaks, record when they recover.
TagPipes is the platform this will be part of.
Link monitoring is still being finished and is not in the product yet. What is in it today is tagging, measurement, consent and privacy tooling.
See what TagPipes does todayFree 14-day trial. No credit card.