The Data Warehouse Toolkit: The Definitive Guid... 💎 🔔

Select the numeric measurements that result from the business process.

These provide the context (the "who, what, where, when, and why") for the facts. Examples include Product , Date , Store , and Customer .

To prevent "data silos," the toolkit introduces the Data Warehouse Bus Architecture . The "glue" of this architecture is Conformed Dimensions —dimensions (like Date or Product ) that are built once and shared across multiple fact tables. This ensures consistency when comparing data across different parts of the business. The Data Warehouse Toolkit: The Definitive Guid...

The "Kimball Method" focuses on integration, making it the practical choice for agile businesses. It stands in contrast to the Inmon Method , which advocates for a top-down, highly normalized "Enterprise Data Warehouse" first. Most modern cloud data warehouses (like Snowflake or BigQuery) are still optimized to run on the star schemas Kimball pioneered.

Add a new column to show the previous value (limited history). Impact on the Industry Select the numeric measurements that result from the

Create a new row with a unique key to track history (the gold standard).

Identify the specific activity to model (e.g., retail sales, hospital admissions). To prevent "data silos," the toolkit introduces the

These store quantitative metrics (measurements) of a business process, such as sale_amount or quantity_sold . They typically contain foreign keys connecting to dimensions.