Why it’s harder than a SaaS-to-SaaS move
Most CRM migrations move data between two SaaS apps with documented APIs. A Siebel move differs in four ways:
- Extraction. Siebel usually runs on-premises or in a private cloud. Data comes out through its integration layer or through database extracts, not a one-click export.
- Data model. Siebel is relational and deep, with many-to-many relationships, child entities and party structures that a SaaS CRM models more flatly.
- Customisation. Long-running Siebel estates often have custom entities, scripted business logic and integrations that grew over many years.
- Volume and history. Years of activities, service requests and attachments.
Oracle still supports Siebel. It states a rolling 10-year support policy on Continuous Innovation releases through at least 2037. So the deadline is yours to set, and you have room to phase the move.
What the native tools do / where they stop
| Need | Salesforce native | Gap |
|---|---|---|
| Read from Siebel | Nothing | You need your own extract |
| Load records | Data Loader (any object, up to 150M records via Bulk API 2.0) | One object per job; lookups built by you |
| Keep dates and owners | Audit fields and inactive-owner permissions, on insert | Users have to exist first |
| Automations | Flow | Siebel workflows and scripts are rebuilt from scratch |
| Coexistence | None | A delta process is needed if Siebel keeps running |
A workable plan
- Inventory: entities, custom fields, record counts per year, attachments, integrations, workflows.
- Target model: map each Siebel entity to a Salesforce object, and decide what collapses (for example, several address entities into one) and what becomes a custom object.
- History cutoff: set one per object. Archive the rest with Siebel IDs intact.
- Pilot: run one business unit end to end, including activities and files, and reconcile the counts.
- Phased loads: each wave upserts on a
Siebel_Row_ID__cexternal ID, followed by a delta. - Process rebuild: build the Flows from the workflow inventory, then test them on the migrated data.
Where FlitStack AI fits
FlitStack AI’s done-for-you service handles migrations like this one: mapping you review, a pilot run, validation and delta sync, then process setup in Salesforce. Scope it on a call, or see Salesforce migrations.