A measure in DAX is basically a calculation that is evaluated dynamically based on the filters and context in a Power BI report. It doesn’t store any value in the model; instead, it computes results on the fly whenever you view or interact with visuals.
For example, in a Sales dataset, if I want to calculate total sales amount, I’d create a measure likeTotal Sales = SUM(Sales[Amount]).
Now, this measure automatically adjusts when I apply filters — like showing total sales by region, product, or year — without needing separate calculations.
In real-world projects, I’ve used measures for key metrics such as revenue, profit margin, and year-over-year growth. For instance, in one of my retail dashboards, I created a measure for Profit Margin usingProfit Margin = DIVIDE([Total Profit], [Total Sales]),
which dynamically recalculated when users filtered by store or category.
One challenge I faced with measures is understanding filter context vs. row context. When I was new to DAX, I sometimes got unexpected results because I used a measure where a calculated column was needed, or vice versa. I learned to handle this by using functions like CALCULATE and FILTER to modify the evaluation context explicitly.
A limitation of measures is that they can’t be used directly in row-level operations — for example, you can’t use a measure in a table relationship or to create another column value per record.
As an alternative, when I need values stored for each row, I use calculated columns; but for anything that depends on user interaction, visuals, or aggregation, measures are always my go-to because they’re lightweight, faster, and keep the model optimized.