Sage 50 and Cin7 integration

What moves between them

The primary flow is Cin7 into Sage 50. After each purchase or invoice in Cin7, ml-connector reads the transaction and posts the GL entry into Sage 50's general ledger, mapped to matching GL accounts and supplier records. In the reverse direction, Sage 50 supplier and customer changes are read via SDK and written to Cin7 to keep master data aligned. Cin7's read-only Chart of Accounts ensures Sage always posts to accounts that exist downstream. Because Sage 50 is pull-only, ml-connector uses Cin7 webhooks where available to detect changes in near real-time, falling back to polling Cin7's API for transaction history.

How ml-connector handles it

ml-connector runs on the Windows machine where Sage 50 is installed, authenticating to Sage via local username and password plus ApplicationID (US) or DataPath (UK). It connects to Cin7 over HTTPS using the provided API key headers. Since Cin7's webhooks are unsigned, ml-connector validates webhook payloads by checking the request origin and query parameters configured in Cin7's UI. On the Sage side, ml-connector reads suppliers and customers from the local SDK and writes GL entries by constructing journal transactions in Sage's data model. On the Cin7 side, it reads purchases and invoices from the REST API and maps them to Sage GL accounts retrieved from the Chart of Accounts. All transactions are tracked with timestamps and retry on failure, and Sage 50's exclusive access requirement is enforced so no interactive user is logged in during SDK polling.

A real-world example

A mid-sized product company runs Sage 50 on a Windows server for accounting and Cin7 Core in the cloud for order management and inventory across multiple warehouses. Previously, every purchase invoice in Cin7 required a finance team member to log into Sage and manually create the matching GL entry, a process that took 30 minutes per day and caused month-end reconciliation delays. With the integration, Cin7 purchase invoices automatically post to Sage 50's GL with the correct supplier and account coding, eliminating manual data entry and reducing the month-end close cycle by one day. Supplier master data stays current: when Sage 50 is updated with a new vendor contact or payment term, it syncs to Cin7 within the hour.

What you can do

• •Post Cin7 purchase invoices to Sage 50 general ledger without manual entry, mapped to the correct GL accounts and suppliers.
• •Sync supplier and customer master data from Sage 50 into Cin7 to keep both systems current.
• •Validate Cin7 webhooks without HMAC signatures using request origin and configured query parameters.
• •Read Sage 50 data via local Windows SDK and write GL transactions that reflect Cin7 purchase activity.
• •Run on the same Windows machine as Sage 50, eliminating the need for remote cloud-to-desktop bridging or third-party ETL tools.

Questions

Why must ml-connector run on the same Windows machine as Sage 50?

Sage 50 exposes data only through a local Windows SDK that requires direct file access to the company data files. There is no remote API, so ml-connector must run on the machine where Sage is installed and authenticate via Windows credentials.

How does ml-connector handle Cin7 webhooks if they are not HMAC-signed?

ml-connector validates unsigned Cin7 webhooks by checking the request's origin IP and verifying query parameters configured in Cin7's UI Automation module. This prevents spoofed payloads while working within Cin7's current webhook design.

What happens if an invoice in Cin7 arrives before the matching supplier is synced to Sage 50?

ml-connector reads Sage 50 suppliers and GL accounts first, creating a mapping table. Cin7 invoices are queued if the supplier or account is not yet present, and the queue is processed after the next sync cycle so no transaction is skipped or posted to an invalid account.