Export Jira Service Management Assets to Excel

Why the normal Jira export doesn't work for Assets

Jira Service Management's Assets module (formerly Insight) stores objects — servers, licenses, contracts, whatever your schema models — as records with custom attributes defined per object type. Jira's standard Excel and Sheets export path was built for issues, not Assets objects, and when it's pointed at an Assets-backed view it leaves the custom-field columns blank. Atlassian documents this gap and tracks it as a known, open limitation, not a bug that's about to be fixed.

That means a CMDB export, an asset-register audit, or a bulk cleanup pass that relies on the standard export silently loses the exact fields — serial numbers, owners, contract dates, custom classifications — that make Assets useful in the first place. The export runs, produces a workbook, and looks fine until someone checks the columns.

What a real Assets export needs

• Direct access to the Assets REST API, not the general Jira export, since that's the only path that returns typed attribute values instead of blanks.
• Support for AQL (Assets Query Language) or at least schema and object-type selection, so the export matches the actual objects you're auditing.
• No artificial row cap — a CMDB or license register is frequently thousands of objects, and a 500-row export limit just means a second manual pass.
• Correct typing on the way out: dates as dates, numbers as numbers, reference attributes as readable labels rather than internal object IDs.

Where this usually goes wrong

Two failure modes show up repeatedly. First, tools that use the general Jira export API against an Assets gadget return the visible columns only (object key and name), with every custom attribute empty, because the export path was never Assets-aware. Second, tools that do hit the Assets API sometimes return reference attributes (an object that points at another object, like a server pointing at its datacenter) as opaque internal IDs instead of the label a human would recognize, which makes the workbook technically complete and practically useless for review. See why Jira Assets reference attributes export as internal IDs instead of names for that second failure on its own.

Checking an export is actually complete

1. Pick an object type with several custom attributes, not just the default Name and Key fields.
2. Export it and open the workbook: every attribute defined on that object type should have its own column, with real values, not blanks.
3. Spot-check a reference attribute (one that links to another object) — it should show a readable label, not an internal ID string.
4. Confirm the row count matches the object count in Assets for that schema and type, with no silent truncation.