Prepare a supported file
Required columns are customer_email, subject, body, created_at, and status. Optional columns are external_id and tags. Use ISO 8601 timestamps with a timezone, and status values open, pending, or resolved. Separate multiple tags with a semicolon. Use quoted CSV fields for message text containing commas or line breaks.
Respect the import limits
Each file can be up to 10 MB and contain up to 5,000 ticket rows. Split a larger migration into multiple files. Attachments, private notes, and threaded message history are not imported; each row becomes one conversation with one historical message. Do not put payment card or sensitive medical data in an import.
Run and review the import
An Owner or Admin opens Settings → Data → Import, uploads the CSV, and reviews the field mapping and validation preview. Invalid rows are skipped and listed in an error report; valid rows still import. Imported conversations do not trigger routing rules, customer emails, or Slack notifications.
Avoid duplicates
If external_id is provided, matching IDs are skipped within the same workspace, including later import files. Without external_id, reimporting can create duplicates. There is no one-click bulk undo. Export and check a small sample before importing the rest; support cannot promise recovery after manual deletion.