Model the Data

A model design with a central fact table linked to surrounding dimension tables via one-to-many relationships, resembling a star.

Measurable, numeric business events (measures) such as sales amount or quantity, plus foreign keys to dimensions.

Descriptive attributes (text, categories, dates) used to filter, group, and label facts, such as Product, Customer, or Date.

It reduces data duplication and cardinality, compresses better in VertiPaq, and lets relationships propagate filters efficiently.

Star schemas have fewer relationships to traverse, simpler DAX, and faster filter propagation; snowflake adds normalized dimension levels that slow queries.

Dimensions are normalized into multiple related sub-tables (for example Product to Category to Subcategory) rather than one flat dimension.