In this classification, one should employ a calculated column and not a measure.
Why Calculated Column?
The evaluated calculated column operates at a row-level evaluation and remains thereby in the dataset. This is because, in classifying an employee's performance (i.e., "Met Target," "Below Target," or "Exceeded Target"), one uses fixed row-level data-sales target versus actual sales. The classification can, therefore, be pre-calculated and stored; thus, it is available for use in filters, slicers, and visuals without recalculating.
Report Performance Impact:
- Performance Benefit: This is a performance shortcut. Since the column values are precomputed and stored, filtering and aggregation actions in visualizations go faster.
- Storage Cost: The calculated column makes the model somewhat larger, but the impact is negligible for small—to medium-sized datasets.
- Measure Limitation: Conversely, if implemented as a measure, the classification would be computed dynamically each time a visual goes through its refresh process, thereby degrading the performance in cases where the datasets become larger.
Conclusion
This suggests that a calculated column is better because it allows efficient categorization, enhances report performance by restricting real-time computations, and facilitates easy filtering and grouping in Power BI reports.
