Sage 50 and Microsoft Power BI integration

What moves between them

ml-connector polls Sage 50 accounts, vendors, customers, purchase invoices, sales invoices, purchase orders, sales orders, and general journal entries on a configurable schedule (typically hourly or after each business event). Each batch of records is transformed to match the Power BI push dataset schema and posted via REST. Historical data is pushed on first sync, then only new or modified records on each subsequent poll.

How ml-connector handles it

ml-connector runs on the Windows machine where Sage 50 is installed, using the local SDK with the company data file path and Windows credentials. It queries Sage 50's modified records by date range or audit trail, transforms vendor names, invoice amounts, GL account codes, and other entities into Power BI table format, and POSTs the rows to a pre-defined push dataset via OAuth2 bearer token. Power BI tokens expire after one hour and are refreshed automatically on each sync cycle. The integration handles the fact that Sage 50 is desktop-only (no remote API) by running the connector process locally, and accounts for GL account constraints by mapping only readable account hierarchies into the push dataset. Retries handle temporary network failures, and audit logs track every record posted to Power BI.

A real-world example

A mid-sized services company runs Sage 50 for accounts payable, accounts receivable, and general ledger. The finance manager and business owner used to export Excel snapshots of vendors and invoice aging every week to email around for decisions. With Sage 50 and Power BI connected, vendor balances, aging buckets, and AP vs AR trends appear in a live dashboard refreshed hourly. The team sees late invoices and vendor payment patterns in real time, month-end close starts with accurate data already in place, and the weekly email ritual is gone.

What you can do

• •Push Sage 50 vendors, customers, purchase invoices, and sales invoices to Power BI in near-real-time for dashboards and reports.
• •Map Sage 50 GL accounts, projects, and bank data into Power BI table schemas so finance can analyze account activity and drill into transactions.
• •Poll Sage 50 on a schedule you set (hourly, daily, or custom) and handle Sage 50's lack of webhooks with reliable polling and retry logic.
• •Authenticate to Power BI with OAuth2 service principal credentials and refresh access tokens automatically on each sync cycle.
• •Run the connector on the Windows machine where Sage 50 is installed, requiring no cloud gateway or complex networking.

Questions

Why does ml-connector need to run on the Windows machine where Sage 50 is installed?

Sage 50 is a desktop application with no cloud REST API or remote integration surface. Integration requires direct access to the Sage 50 SDK and company data files, which are only available locally on the Windows machine. ml-connector runs as a scheduled process on that machine, connects to Sage 50 via the local SDK, and pushes the data to Power BI in the cloud.

Can ml-connector sync in real-time, or does it have to poll Sage 50?

Sage 50 has no webhooks or event system, so ml-connector must poll. The recommended poll interval is 5-15 minutes for near-real-time visibility, or hourly for standard accounting operations. You set the schedule based on how fresh you need the Power BI data to be.

What happens if the Windows machine or Sage 50 connection goes down?

If the connector cannot reach Sage 50, it logs the failure, waits for the next scheduled poll, and retries. Power BI shows the last successfully synced data until the connection is restored. The audit trail in ml-connector tracks every attempt, so you know exactly when Sage 50 was last read and which records made it to Power BI.