Datum
Confluence hygiene · Confluence

How to find broken links in Confluence

Updated September 24, 2026

You almost never go looking for a broken link in Confluence — you find one when a reader clicks it and lands on a missing page. By then the damage is done: the link looked fine, the answer was supposed to be one hop away, and instead there’s a 404. The reason these pile up is that Confluence Cloud gives you no way to find them in advance.

Why there’s no native report

Confluence Cloud has search and the page tree, but neither answers the question “which links in this space point at pages that no longer exist?” There’s no field to search on, and nothing in the UI enumerates a page’s outbound links and checks them. So the native options are all reactive or manual:

  • Wait for the 404. You learn about a broken link when someone hits it — the worst time to find out.
  • Audit page by page. You could open each page, check every link by hand, and note the dead ones. On a space with hundreds of pages this doesn’t happen, so it doesn’t get done.

This isn’t likely to change soon — Atlassian’s own tracker has a link-validator feature request open since 2019, still sitting at “gathering interest” with over a hundred votes.

Links break for ordinary reasons — a page gets moved, renamed, or deleted, or a space is reorganised — and every inbound link to it silently goes dead. Confluence just doesn’t tell you.

Finding them in one pass

The reliable approach is to scan the whole space and check every internal link against the pages that actually exist. That’s what Evergreen does — it walks the space and lists the internal links pointing at pages that are gone, so you can repoint or remove them in a batch, free, on any plan, with no data leaving your tenant.

What that gets you:

  • The dead links in one list — not one 404 at a time in front of a reader.
  • A batch fix — see them together, fix them together, instead of chasing them reactively.
  • A repeatable audit — re-run it as the space grows, so link rot never gets far ahead of you.

Broken links are one half of a space decaying; orphaned pages are the other. A broken link points at a page that’s gone; an orphaned page still exists but nothing points to it. The same reorganisation that breaks a link often orphans a page — so it’s worth catching both in the same sweep. See how to find orphaned pages in Confluence for that side of it.

The takeaway

Confluence won’t hand you a broken-link report — the page tree and search simply don’t answer that question. If your space is big enough that “is this link still good?” is a real worry, you either wait for the 404 or scan the space with a content-hygiene tool like Evergreen and fix the dead links before a reader ever finds them.

Frequently asked questions

Does Confluence have a built-in broken-link checker?
No. Confluence Cloud has no native report that lists broken links across a space — a link-validator feature request has sat open on Atlassian's own tracker since 2019. You discover a broken link when someone clicks it and hits a missing page, one at a time.
Why do links break in Confluence?
Usually because a page was moved, renamed, or deleted, or a space was reorganised. Links that pointed to it — from other pages or references — then resolve to nothing, and Confluence doesn't flag them for you.
How do I find all the broken links in a Confluence space?
Natively you can't in one place — you'd audit page by page. A content-hygiene app scans the whole space and lists the internal links that point at pages which no longer exist, so you fix them in a batch instead of stumbling on them.
What's the difference between a broken link and an orphaned page?
A broken link points at a page that no longer exists. An orphaned page still exists but nothing points to it. They're two sides of the same decay, and a healthy space needs to catch both.

← All Datum guides