Momentum One Academy · lab
Design a Semantic Model for Scale
This lab builds a sales semantic model directly over Fabric Lakehouse Delta tables. The focus is on model shape and reusable calculations: a well-defined star schema, explicit measures, time intelligence, and an alternate date relationship that supports different business questions.
What the lab covers
- Creating a Direct Lake semantic model from a Lakehouse.
- Relating customer, product and date dimensions to sales facts.
- Defining explicit DAX measures and reusable time-calculation patterns.
- Using an inactive ship-date relationship alongside the primary order-date path.
- Enabling query scale-out and reviewing results in a report.
Learning outcome
Participants can explain how semantic-model structure, relationship behavior and reusable measures support consistent analysis at scale. They also practice validating that report visuals respond as intended when users change dates and other filters.
Workshop connection
This overview supports Module 08 — Direct Lake, Semantic Models, Copilot & Data Agents. It connects Lakehouse engineering to a governed business-facing model and introduces the design checks that should precede wider analytical consumption.
Access note
The exercise requires Microsoft Fabric capacity and appropriate permissions. Power BI capabilities may require a paid workspace entitlement or a separate Pro/Premium Per User license. Copilot checks are optional and depend on tenant availability. Detailed instructions and starter materials are provided during the workshop; no participant notebook or data files are published here.
Further learning
Plan your workshop