Back to Topics

Data Modeling

Intermediate
15 min

What is Data Modeling?

Data modeling creates relationships between tables in Power BI, allowing you to build a semantic model that reflects business logic.

Key Components:

  • Connects tables via relationships
  • Creates star schema
  • Enables cross-filtering
  • Improves performance

Why it matters

Connect Data

Link sales to products and customers

Single Source

Create one version of truth

Cross-filtering

Enable interactive reports

Performance

Optimize query speed

Key Concepts

Relationships

Connect tables via common columns

Example: Sales[ProductID] → Products[ProductID]...

Star Schema

Fact and dimension tables

Example: One fact, multiple dimensions...

Cardinality

One-to-many, many-to-many

Example: One product, many sales...

Filter Direction

How filters flow

Example: Single or both directions...

How to use

1

Load tables

Import all needed tables

2

Open Model view

Switch to diagram view

3

Create relationships

Drag between columns

4

Set cardinality

One-to-many typically

5

Configure filters

Single direction usually

6

Validate model

Test with visuals

Example

Goal: Connect sales to products
Drag Sales[ProductID] → Products[ProductID]
Set one-to-many
Single direction filter
Result: Products filter sales, but not vice versa

Pro Tips

  • •Use star schema: Fact in middle, dimensions around
  • •Hide foreign keys: Clean up report view
  • •Mark date tables: For time intelligence

Practice

Build a model linking sales, products, customers, and dates
Next: Dax Functions