In this course, you'll learn how to transform raw, often unoptimised data into a well-designed star schema with fact and dimension tables. You'll see how to use Power Query, Power BI, and Microsoft Fabric (Warehouse and Lakehouse) to build a robust semantic layer that makes reporting simpler, faster, and more reliable. The focus is on practical modelling: from source data to a scalable star schema ready for reporting, data science, and self-service BI.

What you'll learn

• The core principles of dimensional modelling and the star schema.
• The difference between fact and dimension tables and when to use each.
• Transforming raw source data into a star schema using Power Query.
• Designing star schemas for Power BI semantic models and Fabric (Warehouse/Lakehouse).
• How a well-designed star schema reduces DAX complexity and improves performance.
• Guidelines for traceable, reliable, and well-documented datasets.

Programme

Part 1 – Introduction to Dimensional Modelling

• Facts, dimensions, and the importance of the star schema for Power BI and Fabric.

Part 2 – From Source Data to Star Schema

• Analysing source structures and common pitfalls with relational and operational models.

Part 3 – Transforming with Power Query

• Step-by-step transformation of raw data into fact and dimension tables.

Part 4 – Star Schemas in Power BI and Fabric

• Semantic model in Power BI versus modelling in Fabric Warehouse and Lakehouse.

Part 5 – Quality, Performance and DAX

• The impact of a well-designed star schema on DAX complexity, performance, and maintenance.

Part 6 – Gold Datasets and Best Practices

• Designing reusable datasets for data engineers and data scientists, patterns, and Q&A.

• Data engineers looking to design star schemas in Fabric (Warehouse, Lakehouse) or other platforms.
• Data scientists who need reliable, well-structured datasets for analyses and models.
• BI specialists and Power BI developers currently working with non-dimensional sources.
• Data analysts managing gold datasets or semantic models.

Prerequisites

• No specific prior knowledge required; affinity with data is recommended.
• Experience with SQL or building Power BI reports makes the course easier to follow.
• Basic knowledge of data warehousing or modelling concepts is a plus, but not necessary.