Migrate, back up, and organize Confluence content
Confluence content grows without permission. Learn how to export, convert, and migrate it between spaces and formats without losing structure.
Why this matters in Confluence
Over the years a Confluence instance fills with thousands of pages that teams are afraid to touch: there is no safe way to move, convert, or back them up without breaking links, attachments, and page hierarchy. Reorgs demand content moves between spaces, migrations to another platform demand format conversion, and admins need a reliable snapshot before any risky change. The built-in export was never built for any of these — it produces a handful of formats for single pages or whole spaces, with no way to select a subtree, choose a format per audience, or restore from a known-good point.
Native export is limited and inflexible
Confluence’s built-in export covers single pages or entire spaces in a small set of formats — typically PDF, HTML, and XML. There is no way to export a subtree of pages, choose Markdown or JSON for a downstream pipeline, or run a selective archive of one team’s content. Migrations, audits, and tooling integrations all need exactly that kind of granular, cross-format export, which is why admins end up with a folder full of PDFs nobody can reuse.
Content silos form over time
When a team reorganises or splits, the instinct is to duplicate a space rather than move pages, because moving by copy-paste loses child pages, attachments, labels, and history. Six months later there are two copies of the same content in different states of decay, neither trusted. A real migration that preserves the page tree and its metadata is the only thing that keeps a single source of truth.
Cross-format conversion is a real need
A modern docs pipeline is rarely all Confluence. Teams want their wiki pages in Markdown for a static site, in HTML for an internal portal, or in JSON for ingestion by a search index. Converting by hand, page by page, is infeasible past a dozen pages; a bulk converter that handles headings, tables, code blocks, and links is what makes the content reusable outside the wiki.
Backups protect institutional knowledge
A wiki with no reliable content backup is one bad migration, one mistaken bulk edit, or one departing admin away from losing years of institutional knowledge. Confluence page history helps with small mistakes, but it cannot recover from a deleted space or a tenant-level problem. A periodic, inspectable export is the safety net that lets the team operate with confidence.
Storage format editing is the escape hatch
Some problems — a broken macro, a malformed table, a page that will not publish — can only be fixed by editing Confluence’s underlying storage format (XHTML) directly. Without a source editor, admins are stuck filing a support ticket for what should be a five-minute fix. Direct access to the storage format turns the wiki from a black box into something you can actually maintain.
The approach that works
A safe content operation has a consistent shape: export the source to a portable, inspectable format, convert it if the target differs, migrate into the destination while preserving structure, and keep a backup you can restore from. Running those steps with the right formats and tools is what separates a one-evening migration from a multi-week cleanup.
- 1
Define the scope before you touch anything
Decide exactly what you are moving — a single space, a subtree of pages, or a label-based selection — and where it is going. List the page hierarchy, the attachments it depends on, and any cross-space links that will break. A clear scope written down first is what prevents the classic migration failure of discovering halfway through that you forgot a whole child space.
- 2
Export to a portable format
Export the pages you need in a format you can inspect and re-import — JSON for full-fidelity round trips, XML for storage-format preservation, HTML for human-readable archives, or Markdown for downstream docs pipelines. Use JSON or XML when you must preserve every macro and link; use Markdown or HTML when the destination is a different system entirely.
- 3
Convert when the target differs
When moving between systems or formats, convert the export so structure survives the trip. Headings, tables, code blocks, and internal links are the usual casualties — verify each one renders correctly in the target format before you commit to a bulk run. Converting a small sample first catches the formatting bugs that would otherwise repeat across a thousand pages.
- 4
Migrate into the destination space
Import into the target space or a different tool, preserving the page hierarchy instead of rebuilding it by hand. Confirm that parent-child relationships, attachments, and labels came across, and that internal links still resolve. A migration that flattens the tree into a single folder is the most common failure mode, so check the hierarchy explicitly before declaring success.
- 5
Edit storage format when you must
For broken pages or macros that will not behave, open the underlying Confluence storage format directly and fix the XHTML at the source. This is the escape hatch for content problems the editor cannot reach, and it lets you make precise, bulk-safe edits without hunting through the WYSIWYG toolbar.
- 6
Keep a periodic backup
Store a recurring export — weekly for active spaces, monthly for quieter ones — so any future migration, accident, or audit starts from a known-good snapshot. Keep at least one off the live instance, and verify you can re-import from it at least once. A backup you have never restored from is an assumption, not a safety net.
This solution is for you if…
- Admins consolidating or reorganising Confluence spaces after a reorg
- Migration teams moving content between Confluence, another wiki, or a docs pipeline
- Platform and DevOps teams backing up Confluence as part of disaster recovery
- Anyone who needs wiki pages in Markdown, HTML, or JSON for a downstream tool
- Admins fixing broken pages by editing the Confluence storage format directly
Our apps that solve this
Every app is free for up to 10 users on the Atlassian Marketplace.
ConfluenceModern Importer & Exporter for Confluence
Enterprise-grade content migration and backup solution for Confluence
Read the guides
Frequently asked questions
Confluence’s native export is limited to a small set of formats for single pages or whole spaces. A purpose-built import/export app converts pages to and from JSON, XML, HTML, and Markdown in bulk, so you can produce Markdown for a docs pipeline, HTML for an internal portal, or JSON for ingestion — and select a subtree instead of being forced to export an entire space.
Why teams choose NGPILOT
Free for up to 10 users
Every app is free for small teams, with the full trial Atlassian provides for larger ones.
Native Forge apps
Built on Atlassian Forge, running entirely on Atlassian Cloud infrastructure.
Your data stays in your instance
No content is sent to our servers — processing happens inside your Confluence or Jira.
Verified Marketplace vendor
NGPILOT apps are listed, reviewed, and maintained on the Atlassian Marketplace.
Ready to try it?
Every app is free for up to 10 users. Install in a couple of clicks.