Introduction

Semantic models are the foundation of analytics in Microsoft Fabric. They define how data is structured, calculated, and consumed across reports, dashboards, and AI experiences. A model that works for a small team in Power BI Desktop doesn't automatically serve hundreds of users across multiple data stores. When data volumes grow, teams expand, and consumption patterns shift, the design decisions behind the model need to change.

Suppose an organization is scaling its analytics platform in Microsoft Fabric. Their data lives across lakehouses and warehouses, and their existing semantic models were built in Power BI Desktop for small teams. Now those models need to handle larger datasets, more concurrent users, and broader consumption patterns. The models work at their current size, but they weren't designed for scale.

In this module, you make the design decisions that prepare a semantic model for scale. You start by choosing the right storage mode for how data flows into the model. Then you design star schema relationships for clarity and performance. Next, you design calculations that stay performant and maintainable as data volumes and team size grow. Finally, you configure settings that control how the model handles large datasets, concurrent queries, and external tool access.

By the end of this module, you're able to design semantic models that use the right storage mode, follow star schema best practices, include scalable calculation patterns, and are configured for growing data volumes and consumption demands. Models designed for scale also benefit AI consumption, because AI demands the same things from a model: current data, clear relationships, descriptive structures, and the capacity to handle additional query load.