What does grain in a data warehouse fact table define?

• A. The data type of the measures.
• B. The normalization level of dimension tables.
• C. The timing of ETL updates.
• D. The level of detail the fact table records (e.g., per line item, per order, per day).

Answer: (D)

Grain, or granularity, is the level of detail represented by each row in a fact table. It defines what a single row stands for in terms of the measured data. For example, if the grain is per line item, each row captures one line of an order with measures like quantity, unit price, and extended price. If the grain is per order, a row would summarize the entire order, and if the grain is per day, a row would capture totals for a single day. This choice directly shapes what foreign keys to dimensions you include, what aggregations are possible, and how large the table will be. It also guides how you populate the fact table during ETL and how you write queries against it.

The other aspects listed don’t define this level of detail. The data type of the measures concerns how numbers are stored, not what a row represents. The normalization level of dimension tables relates to schema design rather than the detail level of fact data. The timing of ETL updates is about when data is refreshed, not the granularity of the facts themselves.