Oracle Fusion Cloud ERP and SAP SuccessFactors integration

What moves between them

Employee and compensation data flows from SAP SuccessFactors into Oracle Fusion Cloud ERP. New hires, job changes, and compensation updates from SuccessFactors are pulled on a polling schedule and mapped to Fusion cost centers and GL dimensions. Cost center reference data flows in both directions so that SuccessFactors cost codes exist and are active in Fusion before payroll allocations attempt to post to them. Fusion GL transactions and supplier invoices are read-only from the HCM perspective, so the flow terminates in Fusion for financial posting and audit.

How ml-connector handles it

ml-connector stores both credential sets encrypted and handles the SAML assertion signing and key rotation on the SuccessFactors side, refreshing the Bearer token before expiry. On the Oracle Fusion side it accepts the customer-specific pod URL and polls the REST API at a cadence you define, using LastUpdateDate filters to read only changed records since the last run. Employee records from SuccessFactors are mapped to Fusion cost centers and employee dimensions using explicit lookups, so the first sync aligns cost center codes in both directions and subsequent syncs only move new hires and job changes. SuccessFactors enforces a maximum of 10 concurrent threads per client and returns HTTP 429 throttle responses; ml-connector respects the limit and backs off exponentially. Both systems track the full audit trail of every sync run, record mapping, and payload, and failed records can be replayed manually or after credential or dimension mismatches are corrected.

A real-world example

A multinational manufacturing company with plants in three regions runs Oracle Fusion Cloud ERP for finance, procurement, and GL consolidation, and SAP SuccessFactors for global HR and payroll processing. Before the integration, the finance team manually exported employee rosters and cost center assignments from SuccessFactors each month and loaded them into Fusion custom reports to validate GL expense allocations against headcount by department. Plant managers' org changes in SuccessFactors took days to appear in Fusion cost center reports, and month-end close included a full re-reconciliation of labor and allocation accounts because job codes and cost codes did not sync. With Oracle Fusion and SuccessFactors connected, new hires and job moves flow automatically into Fusion cost center dimensions within the same day, and GL expense allocations are validated in real time against the current authorized headcount, eliminating the manual export and load step entirely.

What you can do

• •Sync employee records, job titles, and cost center assignments from SAP SuccessFactors to Oracle Fusion Cloud ERP on a polling schedule.
• •Map SuccessFactors cost center codes to Oracle Fusion GL cost centers and dimensions so payroll allocations are valid on arrival.
• •Align cost center reference data bidirectionally so new cost codes created in either system can be reconciled before financial posting.
• •Authenticate SAP SuccessFactors with SAML Bearer Assertion signing and manage key rotation, and Oracle Fusion with OAuth2 Client Credentials.
• •Track every employee record, job change, and cost center assignment with a full audit trail, with replay capability if downstream GL posting fails.

Questions

Which direction does data move between Oracle Fusion Cloud ERP and SAP SuccessFactors?

The primary flow is from SAP SuccessFactors into Oracle Fusion Cloud ERP. Employee records, job changes, compensation, and cost center assignments move from SuccessFactors into Fusion cost center dimensions for GL allocation. Cost center reference data is aligned bidirectionally so that both systems recognize the same valid cost codes before payroll posting occurs.

How does ml-connector handle the SAML Bearer Assertion requirement for SAP SuccessFactors authentication?

ml-connector stores the RSA private key encrypted and automatically signs SAML assertions for each token request, refreshing the Bearer token before the 24-hour expiry. It also manages the 10-minute SAML assertion expiry window internally, so authentication failures due to key rotation or clock skew are detected and logged before they block a sync run.

Does ml-connector wait for SAP SuccessFactors event push, or does it poll?

ml-connector uses delta-sync polling by default, filtering on lastModifiedDateTime and explicitly requesting $top=1000 records per page to avoid silent record loss that occurs when pagination is not set. Event push can be enabled optionally in the SuccessFactors Admin Center and routed to ml-connector for faster sync, but polling provides a reliable baseline that does not depend on SuccessFactors webhook infrastructure.