Oracle Fusion Cloud ERP and Salesforce integration

What moves between them

The main flow runs bidirectional. Customer records from Oracle Fusion (suppliers and customers) sync into Salesforce as Accounts. Invoices and payments from Oracle Fusion sync into Salesforce as related Invoice records on the matching Account. Opportunities closed in Salesforce flow back to Oracle Fusion as revenue orders on the corresponding customer record. Purchase orders initiated in Salesforce flow into Oracle Fusion for fulfillment. The sync cadence is on a schedule you control, typically every 15 to 30 minutes, filtering Oracle Fusion by LastUpdateDate to retrieve only changed records.

How ml-connector handles it

ml-connector stores both credential sets encrypted and manages OAuth 2.0 token refresh independently for each system. On the Oracle Fusion side it accepts the full pod URL per customer and polls the REST API using OData ?q filter syntax to request only records modified since the last successful poll. On the Salesforce side it uses the My Domain URL and client credentials, handling token expiry after two hours by automatically requesting a new token on the next call. Customer and supplier names are normalized and deduplicated across both systems so a single Account in Salesforce matches a single customer in Oracle Fusion, and invoice lookups use the customer relationship to attach records to the correct Account. Because both systems return timestamps on updates, ml-connector tracks the high-water mark of the last successful fetch, so the next poll starts where the previous one ended and never duplicates records. Every record carries a full audit trail and can be replayed if a downstream call fails.

A real-world example

A mid-market software company uses Oracle Fusion Cloud ERP for financials and procurement, and Salesforce for sales pipeline and customer management. Before the integration, the finance team manually matched new customers in Oracle Fusion to Salesforce Accounts after they were entered, and sales operations exported monthly invoice data from Oracle Fusion into spreadsheets to reconcile against Salesforce Opportunities and closed deals. With Oracle Fusion and Salesforce connected, new customers created in Oracle Fusion appear automatically in Salesforce as Accounts, and the sales team can see the complete invoice history and outstanding balances on each Account without leaving Salesforce. Month-end reconciliation now begins with accurate customer master data, and the manual matching work is eliminated.

What you can do

• •Sync customers and suppliers from Oracle Fusion Cloud ERP into Salesforce Accounts, keeping master data aligned across both systems.
• •Map Oracle Fusion invoices and payments to Salesforce Invoices linked to the corresponding Account.
• •Sync Salesforce Opportunities closed as won into Oracle Fusion as revenue orders on the matching customer record.
• •Authenticate both systems independently with their respective OAuth 2.0 flows, handle token refresh, and poll on a schedule tied to your business cadence.
• •Track every record movement with a full audit trail, timestamp-based change detection, and replay-on-error recovery.

Questions

Which direction does data move between Oracle Fusion and Salesforce?

The sync is bidirectional. Customer and invoice data flows from Oracle Fusion into Salesforce so the sales team has current financial visibility. Closed Opportunities in Salesforce flow back to Oracle Fusion as revenue orders. Both systems have independent credential stores, and ml-connector manages each OAuth 2.0 flow to keep both systems current.

How does ml-connector handle the lack of direct webhooks from Oracle Fusion?

ml-connector polls the Oracle Fusion REST API on a schedule you define, typically every 15 to 30 minutes. It uses OData ?q filter syntax with the LastUpdateDate parameter to retrieve only records changed since the last poll, so polling is efficient and does not re-fetch the entire customer and invoice database on every run.

How are customers and invoices matched across the two systems?

ml-connector normalizes customer names and uses the customer relationship as the primary key for matching. A customer entered in Oracle Fusion is looked up or created as an Account in Salesforce on the first sync, and subsequent invoices from Oracle Fusion are attached to the matching Account. This keeps invoice history and open balances visible in Salesforce alongside the opportunity pipeline.