Last Tuesday a client forwarded me an email from a competitor. Subject line: “Your 404 is costing you traffic.” The competitor had found forty-three dead links across my client’s resource pages and was pitching to replace them. Six weeks earlier I had audited the same site and missed every one. That sting is useful. Broken link building works because the web decays faster than anyone maintains it, and whoever spots the decay first gets the links. The tactic is old, unglamorous, and still one of the highest-conversion outreach plays available. Here is how to run it without burning a month on dead ends.
Broken link building is a link acquisition tactic. You find pages that link to a dead URL, work out what the dead resource covered, create or find a live replacement, then pitch the linking site to swap it in. You are doing the webmaster a favor: they remove a broken experience and restore value for their readers.
Why It Matters in 2026
Link acquisition got harder. Google’s spam systems now discount scaled outreach with a precision that would have looked excessive three years ago. Mass-personalized templates get pattern-matched, ignored, or reported. Broken link building survives because it is structurally different from most outreach: you lead with a defect the recipient owns, not a favor you want.
Second reason: the link graph is rotting. Studies of large crawl samples consistently show a meaningful share of outbound links on older pages point to dead destinations. Sites that were authoritative in 2018 have decayed. Forums closed. Blogs went dormant. That decay is inventory. Every dead link is a page that once earned a citation and now earns nothing — and the linking page still passes authority to its other links.
Ranking impact deserves a caveat. A single replacement link will not move a competitive keyword. What moves rankings is topical authority built from relevant, editorially placed links across a cluster of related pages. Broken link building delivers exactly that when you target pages topically adjacent to your money pages. Treat it as a link volume and relevance play, not a quick win.
Here is the counter-intuitive part: the best broken link targets are not the biggest sites. They are mid-size sites with active editors and infrequent link maintenance. Huge publishers have legal and editorial review that stalls swaps for months. Small blogs have no traffic. The sweet spot is a site publishing regularly, staffed by two to five people, where a broken link is a visible embarrassment rather than an invisible one.
Step-by-Step
You can run this workflow today with a spreadsheet and a free crawler. The sequence matters more than the tools.
1. Pick a topical neighborhood. Choose three to five topics where you already publish strong content. If you sell project management software, your neighborhood is productivity, remote work, and team ops. Do not chase links in unrelated niches just because the broken link is easy to find. Relevance is the entire point.
2. Build a prospect list of linking pages. Find pages that link out to resources in your neighborhood. Use Google search operators: intitle:”resources” + [topic], “useful links” + [topic], [topic] + “recommended reading”. Also pull pages linking to known industry resources with a free backlink checker. Aim for 200 to 500 candidate pages before you start scanning.
3. Crawl for broken outbound links. Run each candidate page through a broken link checker. Free options exist and they work fine at this scale. Export the list of dead URLs along with the page that links to them. You now have a two-column dataset: linking page, dead destination.
4. Filter hard. Most dead links are worthless. Kill anything pointing to a parked domain, a generic homepage, a social profile, or a resource that was never topically relevant. Keep only dead URLs that clearly matched the linking page’s topic. This filter typically removes 70 to 85 percent of your list. That is normal.
5. Check what the dead URL was. Use the Wayback Machine. You need to know what the page contained so you can offer a genuine equivalent. If the dead page was a list of statistics and you have a statistics page, you have a match. If it was a tool and you have a blog post, skip it.
6. Decide: replace or rebuild. Two paths. You either point the editor to an existing page of yours that matches the dead resource, or you create the replacement first. Creating first converts better. “I rebuilt the resource you were linking to” is a stronger pitch than “here is something sort of similar.”
7. Write the pitch. Three sentences. Name the exact page and the exact broken link. State what the dead resource was. Offer the replacement with one line on why it fits. No paragraph about your company. No attachment. No tracking pixel on the first touch.
8. Follow up once, then stop. One follow-up after five to seven business days. If there is no reply, move on. Chasing a third time damages your domain reputation with that editor permanently.
9. Track outcomes by segment. Log response rate, swap rate, and link type. After fifty pitches you will see which page types and which site sizes convert. Double down there and abandon the rest.
Expect a 5 to 15 percent conversion rate on well-filtered lists. On unfiltered lists, expect under 2 percent. The filtering in step four is where the entire tactic lives or dies.
3 Common Mistakes
Mistake one: pitching a broken link without checking whether the page still matters. You find a dead link on a “best tools of 2016” post that has not been touched in eight years and gets no traffic. The editor does not care. Neither should you. Fix: before pitching, check whether the linking page has recent activity, comments, or inbound links of its own. Prioritize pages that still receive traffic. A broken link on a live page is an opportunity; a broken link on a forgotten page is archaeology.
Mistake two: offering a weak replacement. The dead resource was a comprehensive calculator. You offer a 600-word blog post that mentions calculators. The editor sees the mismatch instantly and your credibility is gone. Fix: only pitch when your replacement genuinely serves the same intent. If you do not have a match, build one or skip the target. Half-matches generate polite rejections and burn the relationship.
Mistake three: leading with yourself. “I’m the founder of X and I noticed…” The editor reads that as a pitch and stops reading. Fix: lead with their defect. Open with the URL, the broken link, and what it used to be. The offer comes third. The difference in reply rates between these two openings is not subtle — it is often three to four times.
One more pattern worth naming: sending the same template to every prospect with only the URL swapped. Editors talk to each other, and the swap is obvious. Write the middle sentence fresh each time. It takes forty seconds.
Free Tools
You do not need a paid suite to run this. These free options cover the full workflow.
- Wayback Machine — the single most important tool here. It shows you what a dead URL contained, which is the foundation of every credible pitch.
- Google Search operators — intitle:, inurl:, and quoted phrase searches build your prospect list for free. Nothing paid beats a well-constructed operator string.
- Free broken link checkers — browser-based and CLI options scan a page or a list of pages for dead outbound links and export results as CSV. Sufficient for lists under a few thousand URLs.
- Free backlink checkers — use these to find pages linking to industry resources, which become your candidate list in step two.
- Google Sheets plus a simple status column — your CRM. Prospect, dead URL, replacement, pitch date, follow-up date, outcome. Do not overbuild it.
- Hunter or similar free email finders — limited free credits, enough for a focused campaign of fifty to a hundred pitches.
Skip paid all-in-one outreach platforms until you have proven the workflow manually. Automation amplifies whatever you are doing, including the mistakes above.
Advanced Tip
Most people pitch the editor. Pitch the person who owns the link instead — and give them a reason that is not about your site.
Here is what that looks like in practice. On larger sites, resource pages are maintained by a content manager who inherited them. They did not choose those links and they have no emotional attachment to them. But they do have a quarterly cleanup obligation, and a broken link report is a gift that makes them look diligent. Send a short list of all the broken links you found on their page — not just the one you want replaced — and mention your replacement once, at the end. You are now the person who helped them hit their maintenance target.
The second layer: find pages linking to dead resources where the dead resource had multiple inbound links. A resource that was cited by thirty sites and is now dead is a thirty-prospect campaign in one. Rebuild that resource properly, then pitch all thirty linking pages. This is resource reconstruction, and it outperforms one-off broken link pitches by a wide margin because you are restoring something the web already voted for. That is the difference between borrowing authority and inheriting it.
One constraint: only reconstruct resources you can genuinely improve. A thin rehash of a dead page gets rejected and remembered.
Conclusion
Broken link building rewards diligence over cleverness. The tactic has no secret sauce — find dead links, filter ruthlessly, offer real replacements, and pitch the defect before the favor. What separates people who get links from people who get silence is almost entirely step four. Anyone can crawl a page for 404s. Few people take the time to check whether the linking page is alive, whether the dead resource mattered, and whether their replacement actually serves the same intent.
So here is the question worth sitting with: how many broken links are sitting on the pages that already link to your competitors? That list is free to build and nobody else is working it. Start with fifty candidates this week, filter them honestly, and send five pitches by Friday. The response data will tell you more than any guide can.