Check the external links of a website

Mappi.netMaps of the world, continents, countries and cities
Pages and links | Images and files | Redirections | External links

Your own pages only break when you change something. The pages you link to on other websites break whenever somebody else does, and nobody tells you. This tool crawls your site, collects every link that leaves it and tests each destination, then tells you which ones are dead, which ones have moved and which ones it could not judge.



Analysing the site, this can take a few minutes on a large website. Please leave this page open.

Current status: Initialising...
Pages processed: 0
External links tested: 0 of 0
Problems found: 0

What the report sorts out for you

An outbound link can fail in several ways, and they do not call for the same reaction. The report puts each link in one of three groups and leaves the working ones out, so that the list you get is the list of things to look at.

What we reportWhy it appearsWhat to do
Dead linkThe domain no longer exists, the server refuses the connection, or the page answers 404, 410 or a server error.Replace the link with a working source, point it at an archived copy, or remove it.
Redirected linkThe address still works, but only after one or several redirections to a new address.If the move is permanent, update the link to the final address. If it is temporary, leave it.
Check by handThe site answered 401, 403 or 429, asked for JavaScript or cookies, or did not answer in time. This usually means it refuses robots, not that the page is gone.Open the link in your browser. Most of these turn out to be fine.

Why external links deserve a regular check

Links rot on their own

A page you linked to three years ago had no obligation to stay where it was. Sites are redesigned, companies are bought, articles are moved behind a new address scheme and domains are simply not renewed. The phenomenon has a name, link rot, and it is steady: the older a page is, the larger the share of its outbound links that lead nowhere. None of this shows up in your own server logs, because the failed request never reaches you.

A dead reference costs you the reader's trust

An outbound link is a promise that something useful is on the other side. When a reader follows your source and lands on an error page, the fault is technically somebody else's, and the disappointment is yours. For a tutorial or a reference page, a row of dead sources is the clearest sign that nobody maintains it any more.

An expired domain can become something else

This is the case worth fixing first. When a domain you link to is not renewed, anyone can register it again, and names that already receive links are bought precisely for that reason. Your link then keeps working, with a 200 code, and leads to a page of adverts or worse. An automated check cannot see that the content changed. What it can see is the step before: the domain that stops resolving, which this report lists as dead. Removing the link at that point is what keeps you out of the second stage.

Redirected links are the early warning

A link that now redirects is still doing its job, which is why it is easy to ignore. It is also a link whose owner has already moved once. The redirection will be maintained for a while and then, one day, dropped. Updating the link to the final address while the redirect still tells you where that is costs a minute. Finding the new address after the redirect has gone can cost an afternoon, or prove impossible.

Reading the results

Each entry shows the external address as it is written in your pages, then what happened when it was requested: the error, or each redirection with its code and the final destination. Under it you get the pages of your site that contain the link, up to five of them, because those are the files you open to correct it.

Dead links come first, then redirected links, then links to check by hand. A link that appears on many pages usually sits in a shared part of the site such as the footer or a menu, so one correction clears every occurrence.

Fixing them

For a dead link, first look for the new home of the content: search for the title of the page, or look at the site's own search. If the content has gone for good, the Wayback Machine often holds a copy you can link to instead. If neither exists, remove the link and reword the sentence so that it does not promise a source it no longer has.

For a redirected link, copy the final destination shown in the report into your page. Leave it alone when the redirection is clearly deliberate: a short link, an affiliate or tracking address, or a temporary redirect to a login or consent page.

For a link to check by hand, open it. Many large sites answer 403 to anything that is not a browser, and the page behind is perfectly healthy. The status code reference explains what each of these codes means.

Do not delete a link just because a robot was refused. A 403 or a 429 tells you how the site treats automated requests, not whether the page exists. Check in a browser before changing anything.

For the links between your own pages, use the link checker, and for internal addresses that redirect, the redirect checker. The glossary defines the terms used in the reports.

Common questions

What counts as an external link?

Any link in your pages whose domain differs from the one you entered. The prefix www is ignored, so example.com and www.example.com are the same site. A subdomain such as blog.example.com is treated as external, since it is usually a separate website with its own server.

Are links marked nofollow tested too?

Yes. The attribute is an instruction to search engines, and a reader who clicks the link lands on the same page either way. Internal links marked nofollow are not used to explore the site, in keeping with our other tools, but every outbound link found on an explored page is tested.

Does the tool load the external pages completely?

No. It first asks only for the headers, which is enough to know whether the page answers. When a server refuses that kind of request, it tries once more as an ordinary page request, and stops there. It never follows the links found on the external site.

Why is a link that works in my browser listed for manual checking?

Because the site answered a robot differently from a browser. Protection services challenge automated visitors with a 403 or a page that requires JavaScript, and some servers limit how many requests they accept in a short time. The report says so instead of guessing, and leaves the verdict to you.

Do dead outbound links hurt my rankings?

A few dead links on an old page are a normal part of the web, and search engines know it. What they cost is quality as readers experience it: pages whose sources lead nowhere look abandoned. Keeping them in order is maintenance rather than a ranking trick.

How often should I run it?

Every two or three months for a site that cites many sources, and after any large import of old content. External links change without any action on your side, so the last clean report says nothing about today.