Broken Link Checker
Test every link on a page and see which ones return errors, with the status code for each.
Free and instant — results appear in seconds. No sign-up, no limits, and nothing you type is stored.
Enter a URL and this extracts every link on the page and requests each one, reporting the status code it returns. Internal and external links are both checked, up to forty per run.
It uses HEAD requests, so nothing is downloaded — the check asks for the status and stops there.
A worked example
Checking a resources page with 34 links:
Result: 4 broken
| Link | Type | Status |
|---|---|---|
| /guides/wordpress-setup | Internal | 404 |
| /guides/email-config | Internal | 404 |
| https://oldtool.example.com | External | 0 |
| https://news.example.org/article | External | 404 |
Diagnosing each
The two internal 404s share a prefix, which is the tell. /guides/ was almost certainly renamed — to /tutorials/ or similar — and these links predate the change. Both should be updated to the current path directly.
The 0 response means the host did not answer at all. That domain is likely gone entirely. Worth confirming in a browser, then replacing with an alternative or an archived copy.
The external 404 is a live site that moved or removed the article. A search on the original headline usually locates it, and if not, the Wayback Machine generally has it.
Order of work
Internal first, always. Those cost crawl budget and can orphan pages, and they are entirely within your control. The external ones are worth fixing but they are courtesy rather than urgency.
Why broken links matter, in proportion
A handful of dead external links will not damage your rankings. Google has been clear that outbound links breaking is a normal part of the web ageing, and no penalty attaches to it.
What they cost is trust and traffic. A reader clicking through to a 404 forms an impression about how carefully the page is maintained. Broken internal links are the more serious case: they waste crawl budget, orphan pages that were previously reachable, and often signal a migration that did not finish.
Reading the status codes
404 — the server is up and says this URL does not exist. Genuinely broken. 410 — gone, deliberately and permanently. 500 to 504 — server errors. Often transient, so recheck before removing the link. 403 — access refused. Frequently a bot block rather than a real problem. 0 — no response at all: dead host, timeout, or a server refusing HEAD requests.
The last two need care. Plenty of sites — Cloudflare-protected ones especially — return 403 to anything that does not look like a browser, while working perfectly for real visitors. Always open a flagged link before deleting it.
Where broken links accumulate
Old blog posts linking to tools that shut down. Internal links written before a URL structure changed. Resource pages assembled once and never revisited. Links to news articles that moved behind a paywall or into an archive.
The pattern is age. Anything published more than two years ago is worth a periodic pass.
Fixing them well
For internal links, update to the current URL. Do not rely on a redirect to paper over it — the link should point where it means to point.
For external links, three options in order of preference: find where the resource moved, replace it with a comparable source, or link to an archived copy. Simply deleting the link is a last resort, because it usually leaves a sentence referencing something the reader can no longer see.
Frequently asked questions
Do broken links hurt my SEO?
Broken outbound links are not a ranking penalty — Google treats link rot as normal. Broken internal links matter more, because they waste crawl budget and can orphan pages that were previously reachable. The bigger cost of either is the reader who clicks and hits nothing.
Why does a link show 403 when it works in my browser?
Many sites block automated requests while serving browsers normally, particularly behind Cloudflare or similar protection. A 403 here usually means bot filtering rather than a broken link. Always open it manually before deleting.
Why only 40 links?
Each link is a separate network request, and a page with 200 links would take minutes and hammer other people's servers. Forty covers most pages, and the checker also stops at 25 seconds so a slow host cannot stall the run.
What does a status of 0 mean?
No response at all. Usually the host no longer exists, the connection timed out, or the server rejects HEAD requests specifically. The first is common on links more than a few years old.
Does this check the whole site?
No, one page per run. Site-wide crawling is a heavier job that needs queuing and politeness delays. For a full crawl, a desktop tool like Screaming Frog is the right instrument.
Should I delete broken external links?
As a last resort. First look for where the resource moved, then for a comparable replacement, then for an archived copy. Deleting usually leaves a sentence referring to something the reader can no longer reach, which is its own small failure.
Reviews
No reviews yet. If this tool solved something for you, yours would be the first — and it helps other people decide whether it is worth their time.