Oracle PeopleSoft and Tradeshift integration

What moves between them

The flow runs from Tradeshift into Oracle PeopleSoft. ml-connector polls Tradeshift for new and changed invoices and purchase orders at a schedule you configure, typically once per hour or daily depending on your invoice volume. Each Tradeshift invoice is mapped to an AP voucher in PeopleSoft, matched against your vendor master data, and posted to the appropriate GL accounts and cost centers. Purchase orders flow similarly, creating or updating PO records so your procurement records stay current with what suppliers have received.

How ml-connector handles it

ml-connector stores your PeopleSoft hostname, port, node name, OPRID, and password encrypted, and presents them via HTTP Basic Auth on every REST call to your on-premise or cloud instance. It stores your Tradeshift OAuth 1.0a credentials - consumer_key, consumer_secret, token, and token_secret - encrypted and constructs signed OAuth requests on each API call, including the required X-Tradeshift-TenantId header. On the polling side, ml-connector retrieves documents from Tradeshift using the changedAfter timestamp to fetch only new or modified records since the last run, then maps Tradeshift's UBL 2.0/2.2 XML structure to PeopleSoft AP voucher and PO fields using your configured vendor and GL account mappings. It tracks processed document IDs to prevent duplicate posting, retries failed submissions with exponential backoff if PeopleSoft returns 4xx or 5xx errors, and logs every step so you can audit and replay any record if a downstream workflow fails. Because PeopleSoft requires write operations to go through SOAP component interfaces or Integration Broker async messages rather than direct REST updates, ml-connector composes SOAP calls for AP voucher posting and tracks the natural keys (vendor, invoice number, date) to ensure idempotency across retries.

A real-world example

A mid-sized manufacturing company operates Oracle PeopleSoft for procurement, finance, and payroll, and uses Tradeshift to receive invoices from suppliers across their extended network of 50+ vendors. Before the integration, their AP team received invoices in Tradeshift, downloaded them as PDF or XML, hand-keyed vendor names and invoice amounts into PeopleSoft AP module, and then matched them against purchase orders - a process that took 2-3 days per batch and generated frequent keyboarding errors. With Tradeshift and PeopleSoft connected, invoices land directly in AP as vouchers ready for approval, matched to their source POs and vendor master data, cutting the AP cycle time by 60% and eliminating manual re-entry.

What you can do

• •Post Tradeshift invoices into Oracle PeopleSoft as AP vouchers, automatically matched to vendors and GL accounts.
• •Sync Tradeshift purchase orders into PeopleSoft to keep procurement records aligned with what suppliers have acknowledged.
• •Authenticate PeopleSoft via HTTP Basic Auth against your on-premise or cloud instance, and Tradeshift via OAuth 1.0a.
• •Poll Tradeshift at a cadence you control using timestamp filtering to retrieve only new or changed documents.
• •Track processed documents to prevent duplicate posting, with automatic retries and a complete audit trail on every record.

Questions

Since PeopleSoft is on-premise and Tradeshift is cloud, how does the connector reach both?

ml-connector runs as a cloud service that reaches out to both systems. It connects to your PeopleSoft instance through your configured hostname, port, and node name over HTTPS REST, so your PeopleSoft must be either on the public internet (with IP whitelisting recommended) or exposed via a reverse proxy or VPN. On the Tradeshift side, it is a standard cloud-to-cloud connection using OAuth 1.0a.

What happens if an invoice is already in both PeopleSoft and Tradeshift?

ml-connector tracks the document ID and timestamp of every processed Tradeshift invoice so it does not post the same invoice twice. If Tradeshift redelivers an invoice due to a network issue, ml-connector recognizes it by ID and skips the duplicate post.

Can ml-connector handle both new invoices and invoice changes, such as quantity or amount corrections?

Yes. ml-connector polls Tradeshift using a changedAfter timestamp filter that retrieves all documents modified since the last run, including new invoices and updates to existing invoices. If an invoice amount changes in Tradeshift, ml-connector detects the change and can either post a corrective voucher in PeopleSoft or update the original voucher if your PeopleSoft workflow permits.