Developer tool
Broken Link Checker
Crawl up to 50 pages of your site, check up to 300 links and see every broken link with the page it was found on, with live progress and a CSV export.
Free, no sign-up. Your input is sent to our server only to produce the result and is not stored.
How to use the broken link checker
Find dead links before your visitors do
Broken links frustrate visitors and waste the crawl budget search engines give your site. This checker walks your pages like a visitor would and reports every link that fails, together with the pages that contain it, so you know exactly where to fix it.
How to run a check
- Enter the address of the page to start from. The home page works best because it links to everything important.
- Press Check links. A progress bar shows pages crawled, links checked and broken links found so far.
- When it finishes, read the Broken links list. Each row shows the status code or the error, whether the link is internal or external, and up to five pages where it appears.
- Use CSV to share the list with your team or to track fixes in a spreadsheet.
What it follows
- Links in anchor tags on pages of the same site. The www and non-www versions of a host are treated as the same site.
- Pages are read in full. Other links, images, documents and external sites are checked with a light request, falling back to a normal request if the server dislikes that.
- Mail, phone and script links and in-page anchors are skipped.
Fixing what you find
- 404 on an internal link: correct the link, restore the page, or add a permanent redirect to the closest replacement.
- 404 on an external link: replace it with a current source or remove it.
- Timeouts or host not found: the other site may be down for good. Remove the link.
- Redirected links: update them to the final address to save a round trip.
Limits you should know
Large sites need more than one run. Start from different sections, such as the blog or the shop, to cover them. Pages behind a login are not reachable. Run the check again after your fixes to confirm the list is empty, and repeat it monthly because outside sites change without warning.
Frequently asked questions
How much of my site does it check?
It starts at the page you enter, follows links to other pages on the same site until it has read 50 pages, and checks up to 300 distinct links in total, internal and external. Larger sites get a partial result that says how many links were left unchecked.
What counts as broken?
Links that answer with an error such as 404 or 500, links to hosts that no longer exist, refused connections, certificate errors and requests that take more than eight seconds. Redirects that end on a working page are fine.
What does "could not verify" mean?
Some sites refuse automated visitors and answer 401, 403, 429 or 999 even though the link works in a browser. We list those separately so you can open them yourself instead of flagging them as broken.
Does it run JavaScript?
No. It reads the HTML your server sends, so links that only appear after scripts run, for example in a single-page app, will not be found.
Will the crawl slow my site down?
No. It uses at most four requests at a time and stops after about 100 seconds. Nothing is posted or changed on your site.
Why was my link refused?
The checker only visits public internet addresses. Private networks, localhost and similar addresses are blocked on purpose, and you can run it up to five times per hour.