Oracle Fusion Cloud ERP and Walmart Marketplace integration

What moves between them

Orders placed on Walmart Marketplace flow into Oracle Fusion Cloud ERP as sales orders and then as customer invoices after fulfillment. Walmart seller fees and service charges deduct into Fusion expense accounts mapped to the seller account. Walmart reconciliation reports are read every day and matched against Fusion customer invoices and AR transactions so the settlement payout matches the recognized revenue. Item and inventory data flows from Fusion to Walmart when catalog updates occur, ensuring the marketplace listing reflects current finished goods master data.

How ml-connector handles it

ml-connector stores both OAuth credential sets encrypted and refreshes the Walmart access token every 15 minutes since it expires in 15 minutes, ensuring no request fails due to token age. It presents all required Walmart headers (WM_SEC.ACCESS_TOKEN, WM_SVC.NAME, WM_QOS.CORRELATION_ID) on every call and generates a unique correlation ID per request to track flow through Walmart's system. On the Fusion side, it polls the REST API every 10 minutes filtering by LastUpdateDate to find new and changed orders, and validates each order's customer, GL account, and business unit against Fusion master data before posting. Walmart settlement reports are polled daily, mapped to the sales orders already posted in Fusion, and reconciled line by line to the AR subsidiary ledger. If a Walmart order references a customer that does not exist in Fusion, the record is held in the audit queue and does not post until the customer is added. Webhook events from Walmart are also accepted if enabled, reducing polling latency for high-volume order days.

A real-world example

A mid-sized home goods retailer sources finished goods from multiple suppliers and sells both through their own e-commerce site and through Walmart Marketplace. Before the integration, the finance team downloaded weekly settlement reports from Walmart and re-keyed order and fee data into Oracle Fusion by hand, a process that took two full days and often introduced typos when entering multi-line orders with cross-charged fees. With Oracle Fusion Cloud ERP and Walmart Marketplace connected, each order places immediately in Fusion with the customer detail and line items intact, seller fees post to the correct expense account, and the settlement report reconciles automatically to AR. The finance team now spends those two days on analysis instead of data re-entry, and order-to-cash cycle time dropped by a week.

What you can do

• •Post Walmart orders into Oracle Fusion Cloud ERP as sales orders and invoices with customer, item, and quantity detail intact.
• •Map Walmart seller fees and service charges to the correct Fusion expense and clearing accounts.
• •Reconcile daily Walmart settlement reports to Fusion AR and validate payouts against recognized revenue.
• •Refresh Walmart access tokens every 15 minutes and poll both systems on a schedule aligned to your business hours.
• •Maintain a full audit trail on every record posted, with the ability to replay failed transactions without re-processing the original Walmart order.

Questions

How does ml-connector handle Walmart's 15-minute access token expiry?

ml-connector refreshes the Walmart access token every 15 minutes by calling the OAuth token endpoint with the Client ID and Client Secret before any API call would use an expired token. This prevents 401 errors and keeps orders flowing continuously from Walmart into Fusion without interruption.

What happens if a Walmart order references a customer that does not exist in Oracle Fusion?

ml-connector validates each order against Fusion customer master data before posting it as a sales order. If the customer does not exist, the order is held in the audit queue with a clear error message. Once the customer is added to Fusion, the order can be replayed and posted without re-entering the Walmart order data.

Can ml-connector use Walmart webhooks instead of polling?

Yes, ml-connector can receive Walmart webhooks for order, inventory, and report events if enabled in your Walmart developer account. Webhooks reduce latency on high-volume order days. ml-connector also polls on a regular schedule as a fallback, so orders are never missed if webhook delivery is delayed.