Outbound link checker

Outbound links tell a search engine what your page is about by the company it keeps. Linking to relevant, authoritative sources is a normal part of a well-researched page. Linking to dead or spammy destinations is the problem, and it accumulates quietly as the pages you linked to years ago change hands or disappear.

A standard link passes ranking signal to its destination. Adding rel="nofollow" tells search engines not to pass that endorsement. Use nofollow for paid placements, advertisements and anything a user submitted that you have not vetted. Google treats rel="sponsored" and rel="ugc" as more specific versions of the same idea.

No. It reads the HTML the server returns and extracts every anchor tag in it. A link that a script inserts after the page loads, or one that only appears after a click, will not be in that HTML and will not be listed here. If a page shows far fewer links than you expect, it is probably rendering them client-side.

It requests every listed link from our server, HEAD first and GET if the site refuses HEAD, with a ten-second limit per link, and records the status code. 404, 410, a 5xx or a failed connection is reported as broken. A link that redirected is marked so you can update it to the final address. It stops at 200 links per page.

Because the site refused a request from a server rather than a browser. LinkedIn answers 999 to every bot, X and many others send 403 or 429 from a datacenter address, and some hosts block HEAD outright. Those links usually work fine for a person, so they are flagged for a manual look instead of being called broken. Treating them as broken is how most checkers end up crying wolf.

Any link whose hostname differs from the page you checked, with subdomains of the same domain treated as internal. Anchors to the same page, plus mailto, tel and javascript links, are skipped entirely rather than counted as either.