Sage 50 and Procurify integration

What moves between them

Purchase orders and supplier invoices flow from Sage 50 into Procurify. ml-connector polls Sage 50's purchase orders and vendor invoices by modified date, maps Sage 50 vendors to Procurify vendors, allocates line items to Procurify account codes based on the Sage 50 GL accounts, and posts the records as new purchase orders and bills in Procurify. Reference data (vendors, account codes) syncs bidirectionally so Sage 50 master records create or update their Procurify counterparts, ensuring new orders always reference valid suppliers and GL allocations. Because Procurify order items and payments are read-only and purchase orders are generated from requisitions, ml-connector does not write back to Procurify; the integration flows from Sage 50 outbound.

How ml-connector handles it

ml-connector runs on a Windows machine with Sage 50 installed and authenticated, establishing an SDK session using the ApplicationID, DataPath or CompanyPath, and user credentials from the fact sheet. It polls Sage 50 at an interval you set (typically 15 minutes to hourly) for purchase orders and invoices modified since the last sync, querying by LastModifiedDate or AuditTrail. For each record, it maps the Sage 50 vendor to a Procurify vendor (creating the vendor in Procurify on first encounter), allocates line item GL accounts to Procurify account codes, and constructs a purchase order or bill request. It then authenticates to Procurify using OAuth 2.0 client credentials, refreshing the bearer token on each request cycle, and posts the order or bill. Vendor and GL account writes are idempotent, so repeated syncs do not duplicate records. Because Sage 50 must not have the same user logged in interactively while the SDK session is active, a dedicated integration user account is required. The SDK connection is exclusive, so it must be released between poll cycles to avoid locking the user out of Sage 50 desktop access.

A real-world example

A mid-sized wholesale distributor uses Sage 50 US on their central office Windows machine to manage vendor relationships and invoicing. The procurement team also uses Procurify in the cloud to request supplies, track purchase orders, and manage approvals. Before the integration, purchase orders created in Sage 50 were manually re-entered in Procurify for audit and tracking, and invoices received in Procurify had to be keyed into Sage 50 by the accounting department. With Sage 50 and Procurify connected, each new purchase order in Sage 50 automatically appears in Procurify, and vendor invoices posted in Procurify flow back to Sage 50's accounts payable without re-keying. Vendors are kept in sync across both systems, so procurement staff see the same vendor master and GL allocations whether they look in Procurify or Sage 50.

What you can do

• •Sync purchase orders from Sage 50 to Procurify, mapped to valid vendors and GL account codes.
• •Post supplier invoices from Sage 50 into Procurify as bills, allocating line items to the correct account codes.
• •Keep Sage 50 vendors and Procurify vendors in agreement, creating or updating vendor records as needed.
• •Authenticate Sage 50 using the local SDK with ApplicationID, DataPath, and Windows credentials, and Procurify using OAuth 2.0 client credentials with automatic token refresh.
• •Poll Sage 50 on a schedule you control, with full audit trail on every synced record and replay capability if a downstream post fails.

Questions

How does ml-connector access Sage 50 if there is no cloud API?

ml-connector runs on a Windows machine with Sage 50 installed and uses the native SDK (Sage 50 .NET SDK for US edition, or Sage Data Objects COM/ActiveX for UK edition) to authenticate with Windows-local credentials and read directly from the company data files. The Windows process must have direct LAN access to the data folder and cannot run while a user is logged into Sage 50 interactively.

What happens to Procurify order items and payments in this integration?

Order items and payments in Procurify are read-only through the API, so ml-connector does not attempt to write them back to Sage 50. The integration flows from Sage 50 outbound: purchase orders and invoices posted to Procurify, with vendors and GL accounts synced bidirectionally so the records reference valid master data on both sides.

How does the integration handle Sage 50 exclusive access and Procurify OAuth token expiry?

ml-connector manages Sage 50 exclusive access by releasing the SDK connection between poll cycles, allowing users to access Sage 50 desktop during the polling interval. For Procurify, it refreshes the OAuth 2.0 bearer token on each request cycle, since tokens expire after 24 hours, and it tracks vendor and account mapping to avoid recreating records on retry if a downstream call fails.