Key takeaways
- The mapping and the sign-off usually take longer than the migration run.
- The slower of the two platforms’ API allowances sets the pace of the run.
- Integrations on the old helpdesk share its API allowance with the migration.
- A delta run at cut-over keeps the switch short, so the full run can take as long as it needs.
What sets the pace: API allowances
Every helpdesk limits how many API requests an account can make in a period, and every integration connected to the account shares that allowance. A migration can only go as fast as the slower of the two platforms allows.
- Freshdesk sets its limit per minute by plan, and applies it to the whole account.
- Intercom’s default limit is much higher and is spread over ten-second windows, so writing into Intercom is rarely what sets the pace.
- Zendesk’s Ticket Import API accepts up to 100 tickets per bulk request.
A ticket is also several requests, not one: the record, each reply, its attachments, and often its assignment and state. Attachments add transfer time on top.
How to make a migration faster
- Decide the mapping quickly. Waiting on a decision costs more days than any rate limit.
- Pause heavy integrations on the old helpdesk during the full run.
- Trim the scope: a date range, no spam, no deleted tickets.
- Start the full run early. With a delta run at cut-over, it does not have to finish the night before.
DataGrine migrations can be run self-serve on the platform or fully managed by the migration team, with 24/7 support either way; in both, a delta sync at cut-over keeps the switch short. See how a migration runs.