Conceptual

Power BI Data Modeling: Grain Definition and Impact on DAX Calculations

The core principle defined is "Grain," which serves as a formal definition for determining the specific level of detail that constitutes a single row within a fact table in data modeling domains. This theoretical rule establishes that correct DAX calculations and model integrity depend strictly on identifying whether each row represents an atomic transaction line, an aggregated unit (such as an entire order), or a highly summarized time period (such as a month). Misalignment between the perceived grain of a column's aggregation level and the actual table grain is identified as the primary mechanism causing semantic errors in data analysis.