The workflow before
TEAM Software's Power BI reports depended on data in an AWS RDS application database. Someone exported tables to CSV and imported them into the reporting tools by hand on each refresh. Every step was a chance for a missed file or a broken column, and the process could not support the analytics TEAM wanted to sell to its own customers.
What we built
We built an ETL pipeline in Microsoft Fabric with a medallion design: Bronze holds raw data, Silver holds cleaned data, Gold holds the aggregates the reports read.
An incremental ingestion pipeline now pulls about 20 high-value tables from AWS RDS on a schedule, so only changed rows move.
Before cutover we tested the automated feed against historical reporting data until the numbers matched exactly, so existing Power BI reports kept working with no change for users.
Before and after
| Measure | Before | After |
|---|---|---|
| Data movement to reporting | Manual CSV export and import | Scheduled incremental pipeline |
| Tables automated | 0 | About 20 |
| Refresh effort | About 4 hours of hands-on work a week | None |
| Report accuracy at cutover | Dependent on manual steps | Exact parity with historical reports |
What's next
The Gold layer is the base for analytics TEAM can offer its customers. It is also the clean data a Level 2 reporting assistant needs.
