Table of Contents

1 Dimension Types in Power BI
2 Scenario: Dimension Types in Power BI

When designing a Microsoft Power BI semantic model, understanding dimension types is important for creating efficient and meaningful data models.

This guide defines key dimension types, explains their relevance, and walks through a common scenario to identify the correct dimension type.

Dimension Types in Power BI

There are 4 dimensions types in Power BI as the following:

Degenerate Dimension
Junk Dimension
Outrigger Dimension
Role-Playing Dimension

1) Degenerate Dimension

A degenerate dimension arises when a fact table includes attributes that do not have their own dimension table.

Example: An invoice number stored directly in the fact table without linking to a dimension table.

2) Junk Dimension:

A junk dimension combines unrelated, low-cardinality attributes into a single table to simplify the model.

Example: Combining flags like IsPromotional and IsDiscounted into one dimension.

3) Outrigger Dimension:

An outrigger dimension is a secondary dimension table linked to another dimension rather than directly to a fact table.

Example: A “Location” dimension connected to an “Employee” dimension but not directly to the fact table.

4) Role-Playing Dimension:

A role-playing dimension can serve multiple purposes in the same model by creating relationships with different columns in a fact table.

Example: A date table used for OrderDate, ShipDate, and ReturnDate.

Scenario: Dimension Types in Power BI

You have a Power BI semantic model containing two tables:

DimDate – The main date table in the model.
FactSales – A fact table containing data about sales orders.

In this model, FactSales has three relationships with DimDate:

Active Relationship: Between FactSales[OrderDate] and DimDate[Date].
Inactive Relationships:
- Between FactSales[ShipDate] and DimDate[Date].
- Between FactSales[ReturnDate] and DimDate[Date].

The goal is to determine the type of dimension that DimDate represents in this scenario.

What Type of Dimension is DimDate?

The correct answer is: Role-Playing Dimension.

Why DimDate is a Role-Playing Dimension

A role-playing dimension is a single dimension table that can act in different “roles” within the same data model.

These roles are determined by establishing multiple relationships between the dimension table and the fact table. Only one of these relationships can be active at a time, while the others remain inactive.

How DimDate Functions in This Scenario

Active Relationship: Links the primary FactSales[OrderDate] to DimDate[Date] to filter or aggregate data based on the order date.
Inactive Relationships:
- FactSales[ShipDate] to DimDate[Date]: Used to analyze data based on shipping dates, requiring explicit activation.
- FactSales[ReturnDate] to DimDate[Date]: Used to analyze data based on return dates, also requiring explicit activation.

Activating Relationships in Power BI

Inactive relationships can be activated temporarily using DAX (Data Analysis Expressions) functions like:

DAX Formula Example:

Sales by Ship Date = CALCULATE( SUM(FactSales[SalesAmount]), USERELATIONSHIP(FactSales[ShipDate], DimDate[Date]) )

This flexibility allows users to switch between roles dynamically during analysis without altering the core model structure.

Why Other Dimension Types Do Not Apply

Degenerate Dimension:
- Not applicable because DimDate is not stored in the FactSales table but exists as a separate table.
Junk Dimension:
- Not applicable because DimDate is not a collection of unrelated attributes but a structured date table.
Outrigger Dimension:
- Not applicable because DimDate is directly linked to FactSales, not through another dimension.

Conclusion

In this Power BI scenario, DimDate is a role-playing dimension because it serves multiple purposes (OrderDate, ShipDate, ReturnDate) within the FactSales table.

By understanding and leveraging role-playing dimensions, you can create efficient and flexible data models tailored to complex analytical requirements.

Understanding Dimension Types in Power BI