Migrate from Termene to Twenty CRM

Twenty requires Companies to import before People — Termene's company-person linkage breaks without sequencing

Twenty's CSV import enforces referential integrity: a People record with a companyId foreign key can only import if the referenced Company record already exists in Twenty. Termene stores person records with a link back to their parent company, but if you export both objects simultaneously and load them in the wrong order, every person record lands without a companyId and your relationship graph is severed. We sequence the migration as: (1) Companies first with their source IDs, (2) People second with companyId resolved via Source_ID__c match, (3) Notes and custom fields last. Skipping this sequence is the most common cause of orphaned contact records in a Termene-to-Twenty migration.

Custom fields must exist in Twenty before CSV import — the import creates records, not schema

Twenty's Settings → Data Model requires you to pre-create every custom field before a CSV import can populate it. The import maps column headers to existing fields by name; if a Termene financial indicator like risk_score has no corresponding Twenty field at import time, that column is silently ignored. We generate a custom field creation checklist from the Termene schema audit and deliver it as a pre-migration configuration step. Fields must be created with correct types (Select for pick-list values, Number for numeric indicators, Date for timestamp fields) before the CSV is loaded — changing field types post-import requires record migration, not just field renaming.

Workflow automations and follow-up sequences do not exist in Termene and cannot migrate to Twenty

Termene is a company-intelligence platform with no native workflow, sequence, or automation engine — there are no Termene automations to migrate to Twenty's workflow builder. Reddit discussions about Twenty CRM migration confirm that teams moving from simpler platforms to Twenty face the task of designing their automation logic from scratch for the first time. We provide a Termene audit summary (what events are tracked, what alerts fire, what follow-up actions your team takes manually) as a process reference document for rebuilding that logic in Twenty's workflow builder. This is not a migration of existing automations — it is a process-design exercise your team should plan for post-cutover.

API access for programmatic import is limited to Pro and Organization plans — free self-hosted cannot use REST or GraphQL

Twenty's free self-hosted plan does not include REST or GraphQL API access, webhooks, or MCP server integration. A Termene-to-Twenty migration running on the free self-hosted plan must use CSV export from Termene and CSV import via Twenty's UI (Command Menu → Import records), with a per-export limit of 20,000 records per file. Teams with more than 20,000 Termene records on a free Twenty instance need to split exports by object or upgrade to the Pro plan ($9/user/month) for API-based bulk import. We confirm your Twenty plan tier and export-splitting strategy during the discovery phase.

Duplicate detection on import includes soft-deleted records — restores them if the same unique value reappears

Twenty's import system checks uniqueness against all records, including soft-deleted records visible under Command Menu → See deleted records. If a Termene company record with registration number CUI-123456 is exported, migrated, then later soft-deleted in Twenty, a subsequent delta import with the same CUI value will restore the original record rather than create a new one. We handle this by clearing the uniqueness constraint on Source_ID__c for delta runs if your team has performed manual deletions, or by using a generated migration-specific UUID rather than the Termene ID as the uniqueness key for delta imports. This behavior is documented in Twenty's import documentation and must be planned for in cutover scenarios.