Notion to Confluence

the one decision that matters

Everything written about this migration focuses on the export/import mechanics. Those are the easy part — Atlassian's official importer is genuinely capable and will move tens of thousands of pages, attachments and comments without much drama.

The part that decides whether this migration succeeds is your databases.

Notion databases arrive in Confluence as static tables, with entries as subpages. The rows are all there. Everything that made them a database is not: views, filters, sorts, relations, rollups, formulas, custom properties. Every relational link and every calculated field is gone.

If your team reads documents in Notion, this migration is a good day's work. If your team runs its work out of Notion databases — a sprint board, a CRM, a content calendar with rollups — then you are not migrating a wiki. You are decommissioning a lightweight app, and no import setting changes that.

Decide what replaces each database before you export anything.

what survives, and what doesn't

Comes through cleanly: headings, lists, quotes, dividers, standard formatting, page hierarchy, attachments, and — usefully — comments and original authorship, which the importer can preserve.

Comes through but looks different: to-do lists, info panels and callouts generally keep their content while losing their appearance. Cosmetic, not fatal.

Usually doesn't survive: embedded content (Google Drive, Figma, videos), @person mentions, linked database references and inline page references. These tend to arrive as broken blocks rather than errors, which is worse — nobody notices until someone needs them.

Doesn't survive at all: database views, filters, relations, rollups, formulas. Macros and complex templates aren't supported and can disrupt the destination space if you try.

May not survive: version history. Keep your export archive permanently if you have retention obligations, because the ZIP may be the only copy.

where each database should actually go

Work through your list from step one and sort each database into one of four buckets.

It was really a task tracker → Jira, not Confluence. This is the most common case and the one people get wrong. If the database has statuses, assignees and due dates, Confluence is the wrong destination entirely — you're in the Atlassian ecosystem now, and Jira is the tool that does this. Rebuilding it as a Confluence table produces something nobody updates.

It was a reference table → a Confluence table is fine. Static lists, glossaries, contact directories. These migrate well because they were never really databases.

It was a relational system → budget real time, or reconsider. Multiple databases joined by relations with rollups computing across them have no Confluence equivalent. You're rebuilding, not migrating.

Nobody has opened it in six months → delete it. Migrations are the best opportunity you'll ever get to throw things away, and roughly a third of most Notion workspaces is abandoned. Don't pay to move it.

the step people skip

Keep Notion alive, read-only, for at least a month after you cut over.

Almost every migration horror story is the same story: the import looked fine, the team moved, and three weeks later someone needed a page that got mangled — by which point the Notion workspace had been cancelled to save the seat cost.

A month of overlapping subscription is the cheapest insurance you will ever buy. Downgrade the plan if you must, but don't delete the workspace, and keep the export ZIP somewhere permanent regardless.

should you actually do this

Worth being blunt: most Notion-to-Confluence migrations are not chosen by the people doing the work. They happen because the company standardised on Atlassian, or security wanted everything in one vendor, or Jira was already there.

If that's your situation, this guide is about limiting the damage — and staged imports plus a read-only fallback will limit it well.

If you're choosing freely, read the Notion vs Confluence comparison first. Confluence wins on enterprise permissions, Jira integration and governance. Notion wins on almost everything to do with the daily experience of writing and organising. Moving from Notion to Confluence to get a better wiki is a trade most teams regret; moving to consolidate on Atlassian is a trade that makes sense on its own terms.