ml-connector
IFS CloudAmazon Seller Central

IFS Cloud and Amazon Seller Central integration

IFS Cloud manages manufacturing, finance, and supply chain for your business. Amazon Seller Central lets you sell those products to Amazon customers at scale. Connecting the two keeps your product data, inventory, and customer orders synchronized without manual re-entry. New sales orders from Amazon flow into IFS, and inventory levels update automatically as you fulfill and receive goods. ml-connector handles the different APIs on each side and keeps the two systems in agreement.

How IFS Cloud works

IFS Cloud exposes purchase orders, supplier invoices, GL vouchers, customers, sales orders, inventory, and accounting dimensions through its OData v4 REST API. Each customer has a tenant-specific subdomain (https://<tenant>.ifs.cloud). Authentication uses OAuth2 Client Credentials with a 60-minute token lifetime. IFS does not offer standard webhook subscriptions; instead, it supports Event Actions (server-side triggers) that require manual per-customer configuration in the IFS admin UI. The recommended integration pattern is pull-based polling on OData with filters on modified timestamps. IFS enforces ETag headers for updates and limits result sets to 5000 elements per request, with a rate limit of approximately 1000 requests per minute per tenant.

How Amazon Seller Central works

Amazon Seller Central provides order management, inventory, listings, and financial settlement through the Selling Partner API (SP-API), a REST service with region-specific endpoints. Authentication uses OAuth2, with seller-authorized operations using refresh tokens (valid across sessions) and token lifetimes of 3600 seconds (1 hour). Amazon supports push notifications via EventBridge or SQS but does not offer direct HTTP webhooks; financial settlement events and invoice submissions must be polled instead. Orders older than 2 years are not returned (with exceptions for certain Asia-Pacific regions). Reports API operations are asynchronous: create a report, poll for completion, then download from a presigned S3 URL.

What moves between them

Customer orders and product data flow from IFS Cloud into Amazon Seller Central. Sales orders created in IFS are mapped to Amazon orders, with IFS customer records aligned to Seller Central accounts. Inventory levels are polled from IFS and synchronized with Amazon listings and inventory summaries on a cadence you control (typically daily). Financial settlement and order change events can be received via Amazon SQS or EventBridge when configured, or polled from the Finances API and Orders API. Payment proposals from IFS can be pushed outbound to reconcile against Amazon settlement reports. The main direction is IFS into Amazon, keeping your product catalog and inventory in step with online sales.

How ml-connector handles it

ml-connector stores OAuth2 credentials for both systems encrypted and handles token refresh for each independently when calls return 401 Unauthorized. For IFS, it uses the tenant-specific base URL and OData v4 API, respecting the 5000-element page limit and using concurrent requests with offset-based pagination to avoid timeouts on large datasets. For Amazon, it routes to the correct regional endpoint (North America, Europe, or Far East) based on seller account location and uses the Selling Partner API with refresh token grant for seller-authorized operations. IFS updates require reading the ETag first, then sending it in the PATCH request to enforce optimistic concurrency. Amazon financial events are not available via push, so ml-connector polls the Finances API on a schedule aligned to your settlement cycle. Every record carries source and target IDs for traceability, and all operations include automatic retries with exponential backoff on rate-limit responses (HTTP 429 from IFS, throttling from Amazon). A full audit trail tracks every sync, with timestamps and error details for replay.

A real-world example

A mid-sized consumer goods manufacturer runs IFS Cloud for production, inventory, and order management. They sell branded products directly to consumers through Amazon Seller Central in North America and Europe. Before the integration, the sales team manually created orders in Amazon after receiving them, then emailed product lists to the warehouse team so they could check inventory in IFS. Inventory updates were spot-checked weekly and often out of sync, leading to overselling in low-stock weeks and lost sales from false out-of-stock listings. With IFS Cloud and Amazon Seller Central connected, customer orders flow into IFS automatically, and inventory levels sync daily from IFS to Amazon. The sales team no longer re-enters orders, the warehouse sees real inventory instantly, and overselling is eliminated.

What you can do

  • Sync customer orders from Amazon Seller Central into IFS Cloud as sales orders, with full order line details and customer account mapping.
  • Push inventory levels from IFS Cloud to Amazon Seller Central on a daily or weekly schedule, keeping product availability accurate across regions.
  • Handle OAuth2 authentication and token refresh for both IFS Cloud and Amazon, with automatic retries on 401 and rate-limit responses.
  • Map IFS Cloud customers to Amazon Seller Central merchant accounts and reconcile payment settlements from Amazon against IFS payment proposals.
  • Receive order and inventory change events from Amazon SQS or EventBridge when configured, or fall back to polling if push is not available.

Questions

Which direction does data move between IFS Cloud and Amazon Seller Central?
The primary flow is IFS Cloud into Amazon Seller Central. Product data and inventory levels are pushed from IFS to Amazon on a schedule you define, typically daily or weekly. Sales orders from Amazon are pulled into IFS Cloud as new customer orders. Payment settlements and financial events are read from Amazon and reconciled in IFS. The integration is primarily read-heavy on both sides with writes to IFS on the order side.
How does ml-connector handle IFS Cloud's tenant-specific URLs and OData pagination limits?
ml-connector accepts the full tenant-specific base URL for IFS Cloud (https://<tenant>.ifs.cloud) as part of customer configuration. It respects the 5000-element result-set limit per OData request by using concurrent offset-based pagination for large datasets, fetching multiple pages in parallel to avoid timeouts. ETag headers are read and passed on all update operations to enforce optimistic concurrency control as required by IFS.
Does the integration require Amazon EventBridge or SQS setup for real-time order sync?
Push via EventBridge or SQS is optional. If configured, ml-connector can receive order and inventory change events in near-real-time. If not available, ml-connector polls the Orders API and Finances API on a defined schedule. Financial settlement events and invoice results must always be polled since Amazon does not push them. Polling frequency can be tuned to your settlement cycle and sales volume.

Related integrations

Connect IFS Cloud and Amazon Seller Central

Free to use. Add your credentials, ping your real systems, and see if we fit.

Get started