Guide
Broken link building
The pitch is genuinely elegant: find a link that no longer works on somebody’s page, tell them, offer yours as the replacement. You open with a favour rather than a request, which is why it converts better than a cold ask and why every list of link building tactics still includes it. It also has a flaw sitting in plain sight that almost nobody writes down, and it explains the silence better than any advice about subject lines.
The flaw: a dead link is evidence of neglect
Think about what you actually found. A page is carrying a link to something that stopped existing, possibly years ago. Every visitor who clicked it since then hit an error and nobody changed anything. Whoever built that page is not checking it.
So the signal you used to identify the opportunity is the same signal telling you the email will not be read. The tactic is not broken, but its success rate is decided almost entirely by something the crawler cannot report: whether anybody still tends that page. A dead link on a resource list updated last month is a real opportunity. A dead link on a page whose copyright line says 2019 is a fact about the past.
The filter worth applying first
Before writing anything, look for evidence the page is alive: a recent date, a link added since the dead one, comments, anything. Sorting your list by that is worth more than any improvement to the message, because it is the difference between writing to somebody and writing into an archive.
A 404 is the easy case
Tools that find broken links look for a status code, which catches the obvious failures and misses the interesting ones. A link is useless to that page’s reader, and to that page, in several ways that all return a perfectly healthy 200:
- It redirects somewhere unrelated. The domain lapsed and now belongs to a parking service, or the site was sold, or a migration pointed every old URL at the homepage. The link technically resolves and delivers nothing the author intended.
- It lands on a noindexed page. The destination still exists but has left the index, so it is neither found by search nor passing anything onward.
- The chain ends badly. Three hops later it arrives at an error, or at a login, or somewhere a directive along the way has already cut the chain.
These are better opportunities than plain 404s, precisely because no automated list surfaced them and nobody else has emailed about them. They are also the harder thing to explain in a message, so say what a reader loses rather than quoting a status code.
Every hop, the status at each one, and where it really lands: the difference between a link that works and a link that resolves. Follow one of those links, no account needed.
Check the page before you spend an hour on it
The same minute of work that pays off before any outreach at all, and it applies here twice over, because this tactic tempts you to research a site thoroughly before finding out whether a link from it could have been worth anything.
- Do its outbound links pass anything? Plenty of sites nofollow every external link as policy. That is a fine thing for them to do and it changes what you are asking for: readers, not a link that counts. Read one of its live links.
- Is the page indexable? Old resource pages are sometimes noindexed on purpose, and a link there is invisible to search whatever its rel says. Check it.
- Can it be crawled at all? A path disallowed in robots.txt is never fetched, so the link on it is never read. Read the file.
What makes a replacement a replacement
The dead link did a job for that page’s reader. Your suggestion either does the same job or it is an advertisement wearing a helpful sentence, and the person reading has seen enough of both to tell them apart in about two seconds.
The honest version of this tactic has an uncomfortable consequence: most of the broken links you find are not replaceable by anything you own. Reporting those anyway, with no ask attached, is a real thing to do and occasionally produces more than the pitch would have. Inventing a page afterwards to justify the email is how this became a genre people delete on sight.
The version with the best odds points at you
Somewhere on your own site is a URL that used to exist, earned a link from somebody who thought it was worth citing, and now returns an error. A restructure, a renamed slug, a product page retired, a blog moved off a subdomain. That link is still out there, still pointing, and delivering nothing to anybody.
Recovering it needs no email and nobody’s agreement: a redirect from the old URL to the page that replaces it, and the link works again. This is the same tactic, run against a target you control, with a hundred percent agreement rate. Two things to verify rather than assume: that the redirect lands where you intended rather than on the homepage, and that nothing in the chain has already broken it.
The mirror image is worth stating plainly. Every broken link somebody else is hunting was once a link that worked, and nobody told its owner when it died. That is the same failure, seen from the other end, and it is the end you can do something about.
What we are not going to give you
Reply rates. We do not run outreach as a service and have no data of our own, so any number here would be borrowed from somebody selling the tactic, which is the worst possible source for a figure about how well it works.
Nor a template. A message that obviously came from a file is what taught everybody receiving these to delete them, and the one advantage this approach has over a cold ask is that it can only be written by somebody who actually looked at the page.
Questions people actually ask
What is broken link building?
Finding a link on somebody else’s page that no longer works, telling them, and offering a page of yours as the replacement. It is ordinary outreach with a better opening line: instead of asking for a favour you are reporting a fault, which is why it has always converted better than a cold ask.
Does broken link building still work?
It works where the page is maintained, and most pages carrying dead links are not. That is the flaw built into the tactic: the broken link you found is evidence that nobody has looked at that page in a long time, and the person who would fix it is the same person who has stopped reading the inbox. It works best on pages that are otherwise current, which is a much smaller list than a crawler returns.
How do I find broken links to pitch?
Crawl the resource pages and roundups in your subject and look at their outbound links. The part people skip is checking what dead actually means: a status 404 is the obvious case, but a link redirecting to a parked domain, or landing on a page that is now noindexed, is just as useless to that page’s reader and far less likely to have been noticed by anyone else.
What makes a good replacement page?
Something that does the job the dead page did for that reader. If the missing link was a technical reference and you offer a product page, you are not proposing a fix, you are proposing an advertisement, and the person reading can tell instantly. If you have nothing that replaces it, the honest move is to say so or say nothing.
Is broken link building against Google’s guidelines?
No. It is a person telling another person their page is broken and suggesting an alternative, which is ordinary correspondence. What Google’s spam policies name is buying links, exchanging them at scale, and article campaigns produced to carry them. Nothing about reporting a fault falls into that.
Why does nobody reply?
Usually because the page is abandoned, sometimes because the message reads like a script, and often because the request arrives with no evidence the sender read the page. Nobody outside your own sending can give you a meaningful reply rate, so measure yours and ignore the figures in articles selling the tactic.
What about broken links pointing at my own site?
That is the version with the best odds by a wide margin, and almost nobody does it. Pages of yours that once earned links and now return 404 are links you already own and are throwing away: no email is needed, no stranger has to agree, and a redirect to the right page recovers them. Check that the redirect actually lands where you think, because a chain that ends on an error is the same loss with extra steps.
Every broken link was somebody's working link first.
Yours die the same way: a redesign, a renamed URL, a directory tidying up its listings, and nothing anywhere tells you. ShipDR rechecks every backlink you record on a schedule and reports the day one stops being live, dofollow or indexable, which is the half of this subject you can actually act on.
- Every listing rechecked on a schedule, not once
- Live, dofollow and indexable, read from the page itself
- Told the day a backlink drops or flips
Adding a website and charting its Domain Rating is free, no card. The backlink engine is $10 once, per website, with no subscription.