22-25 April 2026 
Small Team, Medium Data: A Fabric Implementation Journey
Regular 50 minute session for SQLBits 2026Thursday - 01 Jan 1970 - 01:00 - 01:00TL; DR
A candid, step-by-step story of taking Microsoft Fabric from demos to production as a small team in a big org: architecture, governance, capacity, CI/CD, monitoring, and the wrong turns that taught us what works.
Session Details
We’ve all seen the Fabric demos, but what does it really take to make it production-ready?
In this practical session, you'll learn how a small analytics team at a Tier 1 automotive manufacturer took Fabric from early concepts to our main data platform in just two years, including the wins, the setbacks, and the trade-offs behind every decision along the way.
We’ll cover architecture, workspace design, and the essentials of running Fabric at "medium" scale: capacity planning and monitoring, metadata-driven frameworks, adoption and governance, CI/CD and deployment patterns, and our approach to managed self-service BI.
Expect concrete patterns, hard-earned lessons, and a few wrong turns that may help shorten your own path to Fabric in production.
In this practical session, you'll learn how a small analytics team at a Tier 1 automotive manufacturer took Fabric from early concepts to our main data platform in just two years, including the wins, the setbacks, and the trade-offs behind every decision along the way.
We’ll cover architecture, workspace design, and the essentials of running Fabric at "medium" scale: capacity planning and monitoring, metadata-driven frameworks, adoption and governance, CI/CD and deployment patterns, and our approach to managed self-service BI.
Expect concrete patterns, hard-earned lessons, and a few wrong turns that may help shorten your own path to Fabric in production.
3 things you'll get out of this session
- Identify the capabilities you need before your org can call Fabric “in production.”
- Evaluate real architecture and governance trade-offs using a real implementation.
- Learn from our mistakes to avoid common dead ends.
- Evaluate real architecture and governance trade-offs using a real implementation.
- Learn from our mistakes to avoid common dead ends.