Course Outline – Data Modelling With Power BI

Objectives of the Course

After the course, attendees will

  • be able to build more powerful, flexible and accurate dashboards in less time
  • be able to build each dashboard on good data foundations so that they and other can more easily read, understand and maintain them
  • build dashboards and the underlying data model using standard tried-and-tested well-known patterns
  • be familiar with a few common DAX functions that are useful for data modelling

Course Length 2 or days

Course Format

The format is mostly guided lab exercises.  Each lab exercise will start with a short presentation and end with a recap of the main points and a discussion of lessons learned.

Course Pre-requisites

Attendees are expected to have a basic familiarity with Power BI Desktop.  If attendees do not have this, we can arrange a 2-day introduction to Power BI course beforehand.

Attendees are not required to have any previous experience of data modelling – the course will teach this from the ground up.

Attendees will need to bring a laptop with the latest version of Power BI Desktop installed.  Course materials for the lab exercises will be provided for download a few days before the course.

Datasets used in the lab exercises

All datasets for the lab exercises will come from public or open datasets. 

Course Content

Tabular Data 101

Tables, columns and data types.

The grain of a table

Primary and foreign keys

Unique and distinct values

Discrete and continuous variables

Basics of Data Modelling

Tables, columns, datatypes and relationships.  Granularity

Relationships between tables

Fact and dimension tables

Star schemas (and their advantages)

Basic DAX for data modelling

Calculated Columns – and the rare cases when they are useful

Measures – Implicit measures and Explicit measures

Counting items with COUNTROWS()

RELATED()  and its uses

Basic Time Intelligence

More advanced data modelling

Snowflake schemas

Building a data model with several fact tables and common (conformed) dimensions

Data shaping with the Query Editor

Enough data shaping techniques to transform source data into dimension and fact tables.