SYSPRO and Slack integration

What moves between them

Records flow from SYSPRO to Slack. When ml-connector polls SYSPRO OData tables and discovers new purchase orders (PorMasterHdr), supplier invoices (ApInvoice), goods receipts, or general ledger postings (ApPosting, GenMaster), it formats each as a Slack message and posts it to a configured channel via the chat.postMessage API. The same channel routes all event types, or separate channels can be configured per entity (e.g., invoices to finance, purchase orders to procurement). Slack is write-only in this direction; activity does not flow back into SYSPRO.

How ml-connector handles it

ml-connector stores SYSPRO session credentials (operator code, OData password, and server URL) encrypted and logs in to Utilities/Logon to obtain a UserId GUID, cached for the session lifetime and refreshed on timeout. It polls SYSPRO OData at your chosen interval (5 to 15 minutes), filtering ApInvoice, PorMasterHdr, InvWarehouse, and GenBalance on PostDate and InvoiceDate to find new records since the last poll. On discovery, each record is formatted as a Slack message containing key fields such as invoice number, supplier name, amount, and posting date, then posted to Slack via chat.postMessage. Because Slack rate-limits postMessage to one per second per channel, ml-connector queues messages and sends them at throttle. Session token timeouts or logoff trigger an immediate re-authentication, and every message sent carries an audit trail showing timestamp, record ID, and Slack channel.

A real-world example

A mid-sized bearing distributor runs SYSPRO for purchasing, sales, and GL, and uses Slack for day-to-day team communication. Before the integration, the finance team reviewed SYSPRO invoices and postings each morning and emailed the procurement team about new orders from suppliers. The procurement team then checked Slack for the email, switched to SYSPRO, and manually created expedite tasks if a lead time was short. With SYSPRO and Slack connected, every invoice and purchase order appears in Slack within minutes of posting, the team reads the full detail in the message, and procurement can immediately flag expedites without context-switching. Month-end close becomes faster because finance can review all postings in a Slack archive thread rather than running reports.

What you can do

• •Notify Slack channels of new supplier invoices and purchase orders from SYSPRO within minutes of posting, with key fields and amounts in the message.
• •Route SYSPRO events (invoices, postings, goods receipts) to separate Slack channels per entity type or event severity.
• •Authenticate SYSPRO with session tokens and OData credentials, refreshing sessions on timeout without interrupting the flow.
• •Poll SYSPRO OData tables on a configurable interval (5 to 15 minutes) and deliver new records to Slack via message queue.
• •Maintain an audit trail of every record posted, including timestamp, SYSPRO entity ID, and destination channel.

Questions

Which direction does data move between SYSPRO and Slack?

Data flows from SYSPRO to Slack only. SYSPRO purchase orders, invoices, and postings are formatted as Slack messages and posted to channels in real time. Slack is used for notification and team visibility; no data flows back into SYSPRO from Slack.

How does the integration handle SYSPRO session authentication and timeouts?

ml-connector calls the SYSPRO Utilities/Logon method, receives a UserId GUID, and caches it for the session lifetime. If the session times out or is logoff, the next polling attempt detects the timeout, re-authenticates immediately, and continues polling without losing records. Every re-auth and poll carries an audit entry.

What SYSPRO records can be posted to Slack, and how often are they checked?

Purchase orders (PorMasterHdr), supplier invoices (ApInvoice), goods receipts (InvWarehouse), and general ledger postings (ApPosting) are discoverable. ml-connector polls OData tables filtering on PostDate and InvoiceDate at your chosen interval, typically 5 to 15 minutes for financial data, and posts new records within one cycle of discovery.