How to Find Broken Backlinks and Turn Them Back Into Rankings

find broken backlinks

The cheapest links you will ever get are the ones you already earned and then lost. When you find broken backlinks — links pointing at URLs on your site that now return 404, or links on other people’s pages that point to dead resources you could replace — you are working with editorial decisions somebody already made in your favor.

Two distinct jobs share the same name, and they pay off differently. Reclaiming your own broken links is fast, high-yield, and requires nobody’s permission. Hunting broken links on other sites is slower and needs outreach, but it opens doors that cold pitching does not.

Why links break in the first place

Site migrations account for most of it. A CMS change, a move from /blog/post-name/ to /post-name/, a category restructure, an HTTP-to-HTTPS switch done without full redirect mapping — each one can orphan hundreds of inbound links overnight. Product pages get retired. Old campaign landing pages get deleted by someone tidying up. Trailing slashes change.

The equity does not disappear gracefully. A link to a 404 passes nothing. Google will eventually drop the URL from its index, and any authority that page had accumulated is simply gone unless you redirect it. Sites that have been through two or three replatforms often have 10 to 30 percent of their inbound links pointing at dead URLs without anyone noticing.

Find the broken links pointing at your own site

Start here, because it is the fastest win in link building and it costs nothing but an afternoon.

  1. Pull your full backlink profile from a site explorer — every referring domain and the exact target URL each link points to.
  2. Export the list of unique target URLs, deduplicated. A site with 800 links might only have 200 distinct targets.
  3. Run those URLs through a status checker and keep everything returning 404, 410, or a soft 404 (a page that renders but says “not found”).
  4. Sort the dead targets by number of referring domains. The top ten usually account for most of the lost equity.
  5. For each, decide the closest live equivalent and 301 redirect the dead URL to it.

Redirect to the genuinely closest match, not the homepage. A blanket redirect of every 404 to the homepage is treated as a soft 404 and passes little to nothing. If a retired product has a real successor, send it there. If a deleted blog post has no replacement, either restore a version of the content or redirect to the category hub that covers the same topic.

Also check Google Search Console’s crawl stats and the “Not found (404)” report. It shows URLs Google actually tried to fetch, which sometimes surfaces dead targets that third-party crawlers missed. Search Console is ground truth for your own site; third-party indexes are periodic crawls and will always lag.

Check whether your competitors’ dead pages are your opportunity

The mirror image of the exercise: run a competitor’s domain through a site explorer, list their top pages by referring domains, then check which of those URLs are dead. A competitor that shut down a free tool, killed a research report, or migrated badly leaves dozens of live links pointing at nothing.

Every site still linking to that dead page is a prospect with a real reason to hear from you. You are not asking for a favor; you are telling an editor their page has a broken reference and offering a working replacement. That framing is why broken link outreach consistently beats generic pitching on reply rate.

One honest caveat: you need a genuine equivalent. If the dead resource was a 12,000-word original study and your replacement is a 700-word blog post, the pitch fails and should. Build the replacement first, then do the outreach.

Using a broken backlink checker without wasting a week

A broken backlink checker is only as good as the index behind it, and every third-party index is an estimate built from periodic crawling. Two tools will hand you different link counts for the same domain, and neither is wrong exactly — they crawled different subsets of the web at different times.

  • Verify before you act. Open the referring page and confirm the link is still visible in the rendered HTML. Index data goes stale.
  • Ignore links that were already nofollow or from domains with no organic traffic — reclaiming those is busywork.
  • Watch for redirect chains. A link pointing to a URL that 301s to another 301 that ends in a 404 shows as “live” in some tools.
  • Batch by target URL, not by referring domain. One redirect can fix forty links at once.

SEO Rocket’s site explorer covers any domain — yours or a competitor’s — returning referring domains, top pages, anchors, and spam or toxic flags, which is enough to build both lists. The backlink gap view then shows which domains link to rivals but not to you, so recovery work and net-new prospecting come out of the same session rather than two separate projects.

Keep a record of what you checked and when. Link indexes update on their own schedule, so a domain that showed as broken in March may have been fixed by the site owner in April, and re-pitching a link you already recovered is an easy way to look careless in front of an editor.

Prioritize by what the link is actually worth

Not every recovered link deserves the same effort. Rank the dead targets on three axes before you start: how many referring domains point at it, how relevant those domains are to your topic, and whether the page you would redirect to serves the same intent.

A dead URL with 22 referring domains from industry publications is worth an hour of careful redirect mapping. A dead URL with one link from a forum signature is worth nothing. Roughly speaking, the top 10 percent of dead targets carry over half the recoverable value, and clearing them takes a morning.

What to expect after you fix them

Redirects need to be recrawled before anything changes, and that takes time — days for high-traffic pages, several weeks for deep ones. You can nudge it by requesting indexing for the destination URLs and by making sure the redirected pages are linked internally from somewhere crawled often.

The movement, when it comes, is usually modest and broad rather than dramatic and narrow: a lift in impressions across a cluster, a few positions on mid-tail terms, occasionally a page that jumps back to where it sat before a migration. Daily position swings of two or three places are normal noise, so measure the trend across four to six weeks rather than reacting to one check.

Make this a recurring check, not a one-off

Links break continuously. Put a quarterly reminder on the calendar and re-run the same three steps: export backlink targets, status-check them, redirect the dead ones. Ten minutes of the work is the export; the rest is judgment about where each URL should point.

Pair it with a rank tracking view so you can see whether recovered links show up as movement, and with Search Console for confirmation that the redirected URLs are being crawled. SEO Rocket combines both — top-100 snapshots with movement deltas, plus GSC and GA4 connected beside the third-party estimates — so a recovery project has a before-and-after you can actually show a client. Links are necessary, not sufficient; but losing links you already earned is the one avoidable part of that equation.