Migrating a support history, one million tickets at a time
A production support system is a promise: everything customers ever told you stays findable. Changing the system without breaking the promise was the whole job.
§1 The situation
A support organization moving from Zendesk to HubSpot, with years of ticket history that support staff relied on daily. The history couldn't pause, the teams couldn't stop answering customers, and the data couldn't arrive dirty — a migration where operational continuity mattered as much as the transfer itself.
§2 What I did
Led and coordinated the movement of the full support-ticket history: planning the cleanup and validation passes, shaping the workflow changes on the receiving side, defining QA checks, and supporting the production cutover. Python-based cleanup and validation were part of the effort; coordination across teams was constant.
§3 Result
Approximately one million support tickets moved into the replacement support/CRM workflow with cleanup, validation, and QA checks before and during cutover. Support history and operational continuity stayed usable while the team changed its core customer-support systems underneath.
§4 Difficulty & judgment
The hard part wasn't volume — it was preserving data quality while the operational workflows that depended on that data were themselves changing, and sequencing a cutover that stakeholders across systems could absorb. Migration is mostly a coordination problem wearing an engineering costume.
§5 Not claimed
Deliberately left unclaimed
- Sole delivery — this was cross-functional work, and reads that way here.
- An exact ticket count — the one-million figure is approximate and recorded with that qualifier.
- Sole authorship of the Python ETL — my precise hands-on-code boundary is still being confirmed against sources, so it isn't asserted.
§6 The same work, other lenses
- Business systems
- A large CRM and support-system migration coordinated across workflows, validation, QA, and cutover dependencies.
- Operations leadership
- People, data, workflow changes, quality checks, and cutover support for a high-volume operational transition.
- Engineering
- Cleanup and data-quality controls applied at million-record scale under production constraints.