You pull a competitor's backlink profile into your tool of choice, sort by authority, and see a link from a site you'd kill for. You click through, and the page it points to doesn't exist anymore. The referring page is still live. The link is still there. And nobody on your competitor's side has noticed.
That's a broken backlink nobody else is bidding on. Not a broken link on your own site (those you should fix today, not next quarter). A link on somebody else's site, pointing to a dead page on your competitor's domain, that you can redirect or replace.
Most people searching for how to find broken backlinks competitors missed end up with a list of link checkers and a vague instruction to "look for gaps." That's not the workflow. The workflow is more specific, and it lives in a place almost no guide tells you to look: the target URLs of your competitor's backlinks, not the links themselves.
Key Takeaways
- Broken link building starts with dead pages on competitor domains, not with dead links on random sites.
- Standard backlink checkers flag the referring page's status, not the destination page's status — that's why these opportunities get missed.
- Filter competitor backlinks for target URLs returning 404 or 410, then sort by the referring domain's authority.
- Recovery works three ways: a 301 from your own replacement page, a direct outreach email with the 404 as proof, or a guest resource pitch.
- Check for nofollow, redirect chains, and JS-rendered links before you invest outreach time.
Why standard backlink tools miss these opportunities
Here's the thing about how most backlink checkers work. They crawl the web, find a link, and record two things: the page the link sits on (the source) and the page it points to (the target). When you open a competitor's profile, the tool shows you the source URL, anchor text, domain authority, and a couple of flags like nofollow.
What it doesn't do, in most default views, is tell you whether the target URL still resolves. The tool crawled that link months ago. The destination page was fine then. Your competitor deleted it in the meantime, or renamed it, or let it lapse during a site migration. The backlink row in the database still looks healthy because the referring page is still up.
Which means the opportunity sits in a column most people never sort by.
How to check for broken backlinks?
Cross two datasets. First, export your competitor's backlink list with the full target URL, not just the domain. Second, run those target URLs through an HTTP status check — either a bulk checker or a crawler configured to test external URLs.
What you're looking for is a target URL returning 404 (not found) or 410 (gone). A 410 is the better signal: it means the site owner deliberately removed the page and told the server to say so. Either way, a live referring page plus a dead target page equals one recoverable link.
One practical note: URL matching has to be exact. A redirect from /old-post/ to /new-post/ isn't broken — the link still works, it just lands somewhere else. You're hunting 404s and 410s, not 301s. If a page redirects through three hops before reaching a live destination, that's a different (and weaker) situation entirely.
The three blind spots that make these invisible
Tool gaps are real, and they're structural. Here's what I keep running into:
- Nofollow links. Many tools hide or grey out nofollow links by default. A nofollow link to a 404 is still a signal to a human editor who might upgrade it, and plenty of editors do exactly that when you give them a working replacement.
- Redirect chains. If a competitor moved a page and then deleted the redirect, the tool's cached status might still say "200" from an earlier crawl. You'll only catch it with a fresh check.
- JS-rendered links. Links injected by JavaScript don't always show up in a standard crawl, so the tool has no record of them at all. If you suspect a competitor uses client-side rendering, supplement your export with a manual check on their most-linked pages.
None of this means the paid tools are useless. It means the tool gives you the raw substrate, and you do the status check yourself. That extra step is the whole edge.
The workflow, from competitor profile to shortlist
You need three columns: referring page, referring page's domain, target URL on the competitor's site. Export those, and add a fourth that you fill in with the HTTP status.
Step-by-step, in the order that saves time
- Pick two or three competitors — not ten. Two well-chosen ones beat a ten-domain export you'll never finish reading.
- Export their backlinks with full target URLs. Filter out anything pointing to the homepage; homepages rarely die.
- Run the target URLs through a bulk status checker. This takes minutes for a few thousand rows.
- Keep only rows where the status is 404, 410, or "connection timeout" (that last one can mean the domain went dark entirely — check manually).
- Sort by the referring domain's authority, then by whether the referring page looks like an editorial page versus a link directory or a comment.
- Spot-check the top 20 in a browser. Open the referring page, confirm the link is still there, confirm it still points to the dead URL.
Steps one through five can be done in an afternoon. Step six is where you start to see the quality distribution — and where you'll realize that half the high-authority rows are useless because the link is buried in a footer widget on a site that removed human editors in 2019.
How to identify a broken link?
Visually, a broken link on a live page is often invisible — the anchor text still reads normally. You have to click it, or check the HTTP status programmatically.
The reliable method is mechanical: pull the href attribute, request that URL, read the response code. Anything in the 400–499 range is a client error; 404 and 410 are the two you care about. A 403 (forbidden) is worth a glance — sometimes it means the site added bot protection rather than deleting the page, in which case the link isn't broken for human visitors at all.
Same test applies to internal links, if you want to catch your own house in order first. Broken internal links hurt crawl budget and waste the equity you've already earned. Fix those before you go prospecting on someone else's turf — it's a five-minute job with a crawler and it always surfaces something embarrassing.
The comparison: quick check vs. thorough check
There's a trade-off between speed and coverage. Here's how the two approaches compare in practice, based on what each actually catches.
| Approach | What it catches | What it misses | Time for a 3,000-row export |
|---|---|---|---|
| Tool's built-in broken link report | Dead links on the referring page itself | Live referring pages pointing to dead competitor pages | Under 5 minutes |
| Bulk HTTP status check on target URLs | 404s and 410s at scale, including nofollow | JS-rendered links, redirect chains after the first hop | 30–60 minutes including cleanup |
| Manual spot-check on the shortlist | Editorial context, link placement, whether the page is worth the outreach | Volume — you can't do this for 3,000 rows | 2–4 minutes per row |
| Fresh crawl of the referral list | Everything above, plus current status of redirect chains | Anything the crawler is blocked from fetching | Several hours, worth it monthly |
The point of the table isn't that one row wins. It's that the built-in report answers a different question than the one you're asking. You want a live source plus a dead target. The tool's own broken-link feature scans sources.
Recovering the link: three paths, ranked by hit rate
A shortlist of 40 rows might yield 5 to 8 recovered links if you work it properly. Not all will be dofollow, and not all will point to a page that helps you. But that's still a decent afternoon's return, especially compared to cold outreach where a 2 percent response rate is considered normal.
Path one: you own the replacement
If the competitor's dead page covered a topic you've already written about, publish or point a 301 at your equivalent page, then email the referring site's editor with a short note: "Heads up, the link in [article title] points to a page that now returns a 404. Here's a working replacement that covers the same ground: [URL]."
This works best when the referring page is editorial, the site still updates, and your replacement genuinely covers the original topic. Keep the email to four sentences. Screenshot the 404 if you want, but the editor can check in ten seconds — the proof is the point, not the pitch.
I'll admit a mistake here: early on, I sent these emails with a long preamble about the site's content quality. Response rate was near zero. The version that worked was two lines and a link. People who manage editorial calendars don't want a review of their archive. They want the problem fixed.
Path two: a substitute resource pitch
You don't have a direct replacement, but you have something adjacent. This is a softer ask and it converts less often, around one in five of the referrals that respond at all. It's still worth doing for high-authority domains where the link will matter for years.
The framing that works: name the specific dead URL, state what it was about, and offer your page only if it covers the same ground. Don't stretch. Editors notice when the replacement doesn't match the anchor text, and a mismatched substitution gets you flagged rather than linked.
Path three: you need to build the page
Sometimes the dead page was a resource so specific that nothing you have comes close — an old tool, a discontinued dataset, a directory that shut down. Building a page purely to catch one backlink is rarely worth it. Building a page because five referrals point to it is.
That's the real filter: group your 404 shortlist by topic, and if three or more separate referring domains point to dead pages on the same subject, that's a signal you're missing a resource your niche wants. Publish it properly, then reach out to each referring site individually. The page earns its keep twice over — once as a link magnet, once as something readers actually use.
Before you outreach, check for toxicity
A high-authority link to a 404 is not automatically a link you want. Every shortlist needs a toxicity pass, or you'll spend an afternoon chasing links that could hurt you more than help.
How to find toxic backlinks?
Look at the referring domain itself, not the link. The warning signs cluster:
- No editorial content — a page that's a wall of outbound links with one-line descriptions
- Topics unrelated to the dead page's subject
- Do the outbound links on the page point to gambling or adult properties, or to domains that have nothing in common?
- A domain that exists on expired-domain registries, or whose content was clearly swapped wholesale from an unrelated old site
- Sitewide links in a footer or sidebar rather than in the body
Two of those together and I skip the row. One of them and I'll look at the page manually before deciding. The cost of a bad link is asymmetric — recovery takes months, and Google's handling of unnatural links is not something you want to learn firsthand.
How to view broken links?
You need to see the link in context, not just in a spreadsheet. Open the referring page in a browser, find the anchor text, click it. That's the ground truth — it tells you whether the link sits in a 200-word editorial paragraph or a "related resources" widget nobody scrolls to.
For volume work, a crawler pointed at a handful of high-authority referring pages will report every dead link on them in one pass, including ones that point to other sites entirely. Those extra 404s are bonus opportunities: they're on the same page, likely from the same era, and the editor fixes them in the same edit.
What this actually earns you
The pitch for broken link building is that it's cheaper than cold outreach because you arrive with proof of a problem. That's true, with a caveat nobody mentions: you're not the only one who can see the 404. The advantage decays. A dead page on a well-linked competitor domain gets found eventually, and the earlier you check, the cleaner your shot.
Which is the argument for running this pass monthly rather than once. Competitors delete pages. Migrations break URL structures. Sites get redesigned and old links survive the redesign pointing at nothing. The supply regenerates on its own.
The habit that makes it stick is the boring one: keep a running list of competitor 404s you've spotted, and check it against new posts you publish. Sometimes the replacement page you need is one you've already written and simply never matched to the right dead URL. That match, sitting in a spreadsheet, doing nothing — that's the link nobody else picked up.