Broken link building is a way to earn links by fixing other people’s problems. You find a page in your niche that no longer exists but still has other sites linking to it, you check what that page used to say, you publish a better resource on the same topic, and you tell the sites with the dead link that a working replacement exists. The site owner fixes a broken link for their readers. You get a link you earned.
That’s the whole method in four moves. The skill is in choosing which dead pages to chase, and in writing an email a busy editor actually wants to answer. Below is the order I work in, the points where the path splits, and the lines I won’t cross.
One boundary first. If the broken links point at pages on your own site, that’s link reclamation, and I cover it in my backlink audit guide. This post is about dead pages on other people’s sites.
Why Do Dead Pages Still Have Links?
Because almost nobody goes back and checks. A blogger links to a useful guide in 2019, the guide’s owner shuts down the site in 2023, and the link sits there pointing at an error page for years.
The numbers on this are large. Pew Research Center’s study of digital decay found that 38% of webpages that existed in 2013 were no longer available a decade later. The same study found that 23% of news webpages and 21% of government webpages contained at least one broken link.
Ahrefs measured the same problem from the link side. Its link rot study, updated in February 2024, looked at links to 2,062,173 websites and reported that at least 66.5% of links created over the previous nine years were dead. So the raw material for this tactic is everywhere, and it’s one of eight I compare by effort and value in my link building strategies guide. The hard part is finding the few dead pages that are worth your time.
What Decides Whether a Dead Page Is Worth Chasing?
Not every 404 is an opportunity. Before I spend an hour on one, I check four things, and each one changes the plan.
- Topic fit. If the dead page covered something you genuinely know and sell around, continue. If it’s only loosely related, skip it, because the linking sites won’t see your page as a natural swap.
- Who still links to it. Ten links from real, maintained blogs beat two hundred links from scraper sites. If the linkers are mostly junk, stop here.
- Whether you can do better. If you can build a page that’s more current or more useful, continue. If the original was a huge study you can’t match, a thin rewrite will only waste everyone’s time.
- Why it died. A page that moved is a different job from a page that’s gone. I explain both branches further down.
Step 1: Find Dead Pages in Your Niche
There are two ways in. Most of the time I use both.
The backlink tool route. Most paid backlink tools can show a competitor’s pages sorted by how many links they have, with the HTTP status of each page. You filter for pages returning 404 and sort by referring domains. Ahrefs, for example, documents exactly this in its Best by links report: enter a domain, set the HTTP code filter to 404, and look at what still has links. Other tools have similar reports under different names, so check your own tool’s help pages.
The crawler route. Find resource pages in your niche with searches like "your topic" + "useful resources" or "your topic" intitle:links. Then crawl those pages and export every outbound link that returns an error. Screaming Frog’s SEO Spider can do this, and its free version crawls up to 500 URLs, which is plenty for a handful of resource pages.
In my experience, the crawler route finds fewer opportunities, but they’re better ones. A resource page that someone curates by hand is exactly the kind of page whose owner cares about a dead link.
Step 2: Check What Was There With the Wayback Machine
Never build a replacement for a page you haven’t seen. Paste the dead URL into the Wayback Machine and open a snapshot from before the page died.
You’re looking for three things. What the page actually covered, how deep it went, and why people linked to it. Sometimes the answer surprises me: a page I assumed was a guide turns out to have been a single chart or a free calculator, and that’s what everyone cited.
Use the snapshot for research, not as a source to copy. The original text belongs to whoever wrote it. Your job is to cover the same need with your own words, your own examples and fresher facts.
Step 3: Vet the Sites That Still Link to It
Export the list of pages linking to the dead URL and open them. Yes, by hand. A list of 40 linking pages usually shrinks to a much shorter list once you remove scrapers, expired sites and pages that have already removed the link.
I keep a linker only if it’s a real page with a real reader, on a site that still publishes. My full checklist for that lives in my post on how to evaluate backlink quality, so I won’t repeat it here. The short version: if you wouldn’t want the link once you’ve seen the site, don’t email them.
Step 4: Build a Replacement That’s Actually Better
“Better” means better for the reader who clicked that dead link, not longer. Ask what that person needed. Then cover it with current information, clearer structure and anything the original was missing.
A few things I add when they fit: dates that show the page is maintained, a short summary at the top, sources for every number, and a format that matches what people linked to. If they linked to a checklist, build a checklist.
Give the page a sensible URL on your own site and make it useful on its own. If the outreach fails completely, you should still be glad the page exists.
What If the Page Moved Instead of Dying?
Sometimes the Wayback Machine shows the content now lives at a new URL on the same site, and the owner simply forgot a redirect. In that case there’s nothing for you to replace.
What I do: tell the linking sites about the new URL anyway, if I’m already writing to them about something else, and don’t pitch my own page. Honestly, it feels odd to help a competitor. But pretending a live resource is dead is a lie, and editors remember who wasted their time.
What If You Already Have a Matching Page?
Then skip Step 4, but check your page against the archived one first. If the dead page covered five subtopics and yours covers two, close the gap before you send a single email.
This branch is the fastest win in broken link building, because most of the work is already done. It’s also where people get lazy and pitch a page that only half matches. If your page isn’t a fair replacement for what the reader expected, the editor will see that in ten seconds.
Step 5: Send an Honest Outreach Email
Keep the email short. Lead with the broken link, because that’s the useful part for them. Mention your page second, and make it easy to say no.
Here’s the template I use. Copy it, then change every bracket.
Subject: Broken link on your [page topic] page
Hi [first name],
I was reading your page on [topic] ([URL of their page]) and noticed the link
to [name of dead resource] now goes to an error page. It's in the section about
[section or sentence].
The original seems to have been taken down. I wrote a guide that covers the same
ground, updated for [month year]: [your URL]. If it fits, it could work as a
replacement. If not, you may still want to remove or swap the dead link.
Either way, thanks for keeping the page up.
[Your name]
[Your role, company]
What’s not in it matters too. No fake compliments about an article you didn’t read, no “as a loyal reader,” no offer to pay or swap links. If an editor replies asking for money, my answer is no. Google’s link spam policy treats buying links for ranking purposes as spam, and any paid link has to carry rel=”sponsored” or rel=”nofollow”.
I send one follow-up after about a week, then stop. Two emails is polite. Five is spam.
What Results Should You Realistically Expect?
I can’t give you a response rate, and I’d be suspicious of anyone who quotes one as a rule. Replies depend on your niche, how well the linking sites are maintained, how good your replacement is, and whether the email reached a person who can edit the page.
What I can tell you is the shape of it. Many emails get no reply at all. Some editors fix the link by removing it rather than swapping it in your favor. A few say yes. So track your own numbers from the first campaign: emails sent, replies, links placed. After 30 or 40 emails you’ll know far more about your niche than any industry average could tell you.
Broken link building is also slow. Finding, vetting, writing and emailing takes real hours, and the payoff comes in single links, not batches.
Where Is the Ethical Line?
Broken link building is one of the cleanest outreach tactics I know, because the core of it is helpful: you point out a real problem. It stays clean only if the rest of the process is honest too.
My rules:
- Never report a link as broken without checking it yourself, in a normal browser.
- Never copy the archived content and present it as your own.
- Never pay, swap links or offer anything for the placement.
- Never automate mass emails to thousands of linkers you haven’t looked at.
- Respect a “no” and a no-reply. Don’t keep chasing the same editor.
If you want the wider context on where tactics cross Google’s line, my post on white hat vs black hat link building walks through it.
Edge Cases That Change the Path
- The dead page was a tool or calculator. If you can’t build a working equivalent, skip it. A blog post isn’t a fair replacement for a tool.
- The linking page is itself abandoned. If the last update was years ago and the site has no contact route, move on. Nobody will edit it.
- The dead page belonged to a direct competitor. It’s fine to pitch, but be extra careful that your replacement is genuinely better, and never criticize the competitor in your email.
- The topic is medical, legal or financial. Your replacement needs named expert sources and a clear author. Editors on these pages are rightly picky.
Broken Link Building Decision Matrix
| If the dead page | And the linkers are | Then |
|---|---|---|
| Matches your topic and you have no page yet | Real, maintained sites | Build a better replacement, then email |
| Matches a page you already have | Real, maintained sites | Close any content gaps, then email |
| Moved to a new URL on its own site | Any | Don’t pitch your page; mention the new URL if you’re writing anyway |
| Was a tool, dataset or study you can’t match | Any | Skip it |
| Matches your topic | Mostly scrapers or dead sites | Skip it |
| Is only loosely related | Any | Skip it |
Who Should Do This Work?
If you know your topic and can write well, you can run broken link building yourself. It needs patience more than money.
What it can’t do is fix a site with no basic presence. If your business has no profiles, no listings and no niche directory entries, editors who check you out will find very little. That foundation is what my basic link building service covers: brand profiles, business listings and niche directories, nothing more. Broken link outreach works better once that layer exists. If you’re not sure where your site stands, a free SEO audit will show you. And if you’re after other zero-cost tactics, my guide on how to get backlinks for free compares them by effort.
Frequently Asked Questions
Is Broken Link Building Still Worth It in 2026?
Yes, for sites that can produce a genuinely useful replacement page. Link rot keeps creating new opportunities, as the Pew and Ahrefs studies above show. It’s slower than people expect, though, and it works best in niches where people still maintain resource pages by hand.
Is Broken Link Building White Hat?
It is when you do it honestly. You point out a real broken link, offer a relevant replacement, and accept whatever the editor decides. It stops being white hat the moment you pay for the placement, swap links, or misrepresent a page as broken.
How Do I Find Broken Backlinks on Competitor Sites?
Use a backlink tool’s report of a competitor’s pages, filter for pages returning a 404 status, and sort by referring domains. Then open the Wayback Machine to see what each dead page contained before you decide whether to build a replacement.
How Many Follow-Up Emails Should I Send?
One. Send it about a week after the first email, keep it to two or three lines, and then stop. If the editor doesn’t reply after two messages, the answer is effectively no, and more emails only damage your reputation.
Last updated: October 2026 by Mizanur Rahman
