Migrating off HubSpot without losing a quarter of pipeline data

CRM migrations fail quietly. Nobody loses all their data at once — they lose 20% of activity history, a handful of custom field mappings, and every automation nobody remembered to rebuild. By the time it's noticed, forecasting is already broken for a quarter.

Before you touch the export button

• Audit what you actually use. Export every custom field, and mark which ones are load-bearing (used in reports, automations, or filters) versus dead weight nobody's touched in a year.
• Map fields before you migrate data, not after. A field mapping done post-import means re-doing the import.
• Decide what "historical" means. Do you need every email ever logged, or just the last 12 months of activity? Full history migrations take longer and aren't always worth it.

The migration itself

1. Export contacts and companies first — they're the foundation everything else attaches to
2. Import deals next, mapped to your new pipeline stages (not a 1:1 stage copy — this is your chance to fix a bloated pipeline)
3. Migrate activity history (calls, emails, notes) last, since it depends on contacts and deals already existing in the new system
4. Run a parallel period — both systems live for 1-2 weeks — before fully cutting over

What breaks if you skip this order

Automations tied to old field names silently stop firing. Deals imported without their original close dates distort your historical win-rate reporting. Email history imported before contacts exist gets orphaned and never resurfaces. None of these throw an error — they just quietly make your new CRM's data untrustworthy from day one, which defeats the entire point of migrating.

The fix isn't more tooling. It's sequencing the migration in the order data actually depends on itself.