How Syntra ETL transformed and loaded the complete supplier master — twelve related SRM objects, from supplier headers and sites to bank details, business classifications, category hierarchies and attachments — from Oracle EBS into Oracle Fusion across 18 business units.
Supplier data is deceptively deep in Oracle Fusion — a single “supplier” spans a dozen linked objects that must load in the right order:
Category type, class and the L1–L5 category hierarchy were incomplete for many suppliers, and several business-classification fields were absent. Each had to be sourced or defaulted before Fusion would accept the record.
Business classifications had to carry the Fusion lookup code (e.g. the coded value) rather than descriptive text — a common silent cause of import failure.
Bank and branch names differed only by case or spacing between source and Fusion (for example “Main branch” vs “MAIN BRANCH”). These were normalized and matched to the Fusion bank masters.
A site must be active, carry the purchasing purpose, be tied to the procurement BU and hold a valid assignment for the requisitioning BU — validated up-front instead of failing at load.
Because master data was loaded across multiple mock cycles, batch keys prevented “already exists” duplicates when the same suppliers were reloaded.
Supplier documents were loaded to Oracle UCM with MIME, size and parent-linking checks so each file attached to the correct supplier.
Mandatory-field, datatype, lookup and referential-integrity checks against the Fusion targets ran in staging, so exceptions were reported back for correction before any Fusion import.
Category types, classes, the L1–L5 browsing hierarchy and classification codes were mapped to Fusion value sets in the tool.
Supplier sites were assigned across the in-scope business units using the encoded 18-BU logic.
A small golden-record set proved the mappings; full-volume runs followed once the profile load was clean.
Source → staged → Fusion counts plus row/column checks were produced for each supplier object.
All of them: headers, addresses, sites, site-to-BU assignments, contacts, browsing categories, products & services, business classifications, banks, tax categories and attachments.
Most failures trace to master-data gaps and target mismatches — missing category hierarchy, classification text instead of codes, inactive or mis-assigned supplier sites, and bank-branch name mismatches. Syntra catches these in load-readiness checks before import.
Batch keys and de-duplication made re-runs idempotent, so reloading the same suppliers across mock cycles did not create duplicates or “already exists” errors.
Tell us your source EBS footprint, modules and business units — we'll scope a mock-run-based transform-and-load plan on the Syntra ETL platform.