Link Rot: Why Links That Worked at Publish Time Quietly Stop Working, and How to Catch It

A link extractor showing 200 links on a page tells you the links exist — it doesn't tell you whether 30 of them lead to pages that no longer exist, and "link rot" accumulates silently because broken outbound links don't produce any visible error on the page that contains them

The previous articles on this site covered link auditing for internal linking strategy, external/toxic link audits, and link attributes (nofollow/sponsored/UGC). This article addresses link rot — the gradual accumulation of links pointing to pages that have moved, been deleted, or otherwise stopped existing — and why it's a maintenance problem that's easy to overlook precisely because it's invisible until someone clicks.

Why link rot happens, and why it's continuous, not a one-time event

Every link on your site points to some URL — on your own site, or on someone else's. Links to your own site can break if you restructure URLs (covered in previous slug/redirect articles) — but links to external sites break for reasons entirely outside your control: the external site restructures their URLs, shuts down, gets redesigned, or the specific page is removed.

This means: a page that had zero broken links when published can accumulate broken links over time, without anyone touching that page — the page's own content/links haven't changed; external sites have. A 5-year-old article with 20 external links might have a meaningfully different "broken link count" today than it had when published — and this degradation is continuous and ongoing, not a one-time event to "fix once and forget."

Why broken links are invisible without specifically checking

A broken link doesn't look broken — to a casual reader scanning a page, <a href="https://example.com/article">this article</a> looks identical whether example.com/article currently exists or returns a 404. The "brokenness" only becomes apparent when someone clicks — and even then, many users, encountering a broken link, simply go back without reporting it — site owners rarely hear about broken links from users, even when users encounter them regularly.

This is fundamentally different from, say, a visual rendering bug (a broken image, a layout issue) — which is visible to anyone viewing the page, without needing to interact with anything. Broken links require active checking (either automated, or via someone clicking every link) to discover — there's no passive signal.

User experience and SEO implications

For users: clicking a broken link is a small, but real, negative experience — particularly frustrating if the link was presented as a valuable resource ("for more detail, see this guide") and turns out to lead nowhere.

For SEO: while a single broken outbound link is unlikely to have a large, direct ranking impact — a page (or site) with many broken links can contribute to an overall impression of "this content isn't well-maintained" — and, more concretely, broken links represent "dead ends" in a page's link structure — any "link equity"/relevance signal that might have flowed through that link (to the external resource, or, if it's an internal link that's broken, to the intended internal page) is, effectively, wasted — pointing at nothing.

Checking for broken links: a recurring, not one-time, task

Given that link rot is continuous — checking once, when a page is published, doesn't address links that break afterward — periodic, recurring checks are needed to catch newly-broken links.

For a small number of key pages (a site's most-trafficked or most-important content) — periodic manual re-checking (using a link-extraction tool, then spot-checking a sample of the extracted links) might be feasible.

For larger sites, dedicated broken-link-checking tools (which crawl a site, extract all links, and check the HTTP status of each — flagging 404s and other error statuses) are more practical than manual checking — running such a check periodically (monthly, quarterly — depending on site size and how quickly link rot tends to accumulate for your specific outbound-link patterns) provides an ongoing "here's what's currently broken" view, rather than a one-time snapshot that becomes stale.

Fixing broken links: several approaches, depending on why it broke

The external page moved (URL changed, but content still exists somewhere): update the link to the new URL — if findable (the external site's own search, or a web search for the content's title/topic, might locate where the content now lives).

The external page is gone entirely (the resource no longer exists anywhere): options include: removing the link (while keeping the surrounding text, if it still makes sense without the link); replacing the link with a link to a different, similar resource* (if the original was illustrative, a comparable alternative might serve the same purpose); or, for historical/archival purposes, linking to an archived version (via web-archive services that preserve historical snapshots of pages — though linking to archived versions of content that no longer exists live has its own considerations, and isn't always the most useful experience for a reader, depending on context).

The external page now redirects somewhere unrelated (a domain was sold/repurposed, and now redirects to something entirely different from what was originally linked) — this is arguably the worst case: the link technically "works" (returns a 200, via the redirect) — but leads to content entirely unrelated to what was intended — a "works but wrong" link, which automated "check for 404s" tools might not flag at all (since the final destination returns a successful status code — it's just not the content that was originally being linked to) — catching this category generally requires some human review (spot-checking where links actually lead, not just whether they "work" in an HTTP-status-code sense).

How to use the Link Extractor on sadiqbd.com

1. Extract all links from a page, then check the status of each — flagging 404s and other error responses as candidates for the fixes described above
2. For older content specifically: prioritize checking older pages first — given that link rot accumulates over time, older content is statistically more likely to have accumulated broken links than recently-published content
3. Establish a recurring check schedule — rather than treating link-checking as a one-time cleanup task, recognizing it as an ongoing maintenance activity, similar to other recurring site-health checks

Frequently Asked Questions

Is there a way to prevent link rot, rather than just detecting it after the fact? Not entirely — link rot is, fundamentally, external sites' behavior, which you don't control. Some partial mitigations: linking to stable, well-established sources (major institutions, long-running publications) that are statistically less likely to disappear/restructure than smaller/newer sites — though even major sites do restructure URLs occasionally. For critical references, some publishers create their own archived copy of external content at the time of linking (with appropriate attribution), reducing reliance on the external site continuing to host that exact content at that exact URL — though this raises its own considerations (copyright, staying current if the original is updated) that go beyond the scope of "just checking for broken links."

Is the Link Extractor free? Yes — completely free, no sign-up required.

Try the Link Extractor free at sadiqbd.com — extract every link on any page, ready for status-checking and link-rot audits.