A measure is a Power BI calculation, written in DAX, that produces a result on demand and recalculates as a report is filtered. A calculated column is also written in DAX, but it adds a fixed value to every row of a table and stores it in the data model.
The same language, two different jobs
Measures and calculated columns are the two everyday ways to create a calculation in Power BI. Both are written in DAX, the formula language Power BI uses, and every DAX formula starts with an equals sign. That shared starting point is why people mix them up.
The difference is not the language; it's where the result lives and when it's worked out. A calculated column is part of a table. A measure is a standalone calculation that runs only when a report needs it. Get that distinction right and the rest of DAX gets much easier.
What a calculated column is
A calculated column is a new column you add to an existing table, with a DAX formula that sets the value for every row. Microsoft's calculated columns tutorial gives a typical example: a table has a City field and a State field, and you want a single Location column reading "Miami, FL".
The value is worked out once, row by row, and then stored in the data model. It is recalculated only when the data is refreshed. Because every value is held in the model, a calculated column adds to the file size and the memory the report uses. You use a calculated column where you would use any other field: to group, filter, sort or place on an axis.
What a measure is
A measure is a calculation that has no fixed result. Microsoft's DAX overview describes measures as "dynamic calculation formulas where the results change depending on context". A Total Sales measure shows the right total for whatever the report is filtered to, whether that is one product, one region or everything.
A measure stores no per-row values at all, only its formula. It is worked out the moment a report uses it, and again every time someone changes a filter or a slicer. That makes measures light on the model and ideal for any figure that should react to what the reader is looking at: totals, averages, percentages and comparisons over time.
Measures vs calculated columns: side by side
The two differ on four things that matter in everyday report building:
| Calculated column | Measure | |
|---|---|---|
| Where the result lives | A new column inside a table | A standalone calculation, in no table's rows |
| When it is worked out | Row by row, when the data is refreshed | On demand, each time a report uses it |
| Storage in the model | A value is stored for every row, adding to file size | Only the formula is stored; no per-row values |
| What it is for | A new field to group, filter or sort by | A figure that reacts to filters: totals, averages, trends |
There is a deeper reason behind the table. A calculated column is evaluated in "row context", meaning it sees one row at a time, so its value is fixed. A measure is evaluated in the report's filter context, so its value moves with the report. That single idea, called context, is the heart of DAX.
Which one should you use?
A useful default: reach for a measure first, and use a calculated column only when you genuinely need a per-row value. Measures are lighter on the model and cover most of what a report needs.
Choose a measure when the answer should change as the report is filtered: a sum, an average, a count, a percentage of total or a year-on-year figure. Choose a calculated column when you need a new attribute on each row to build the report around, such as joining a year and a quarter into a "2026 Q1" label, or sorting customers into spend bands. If you find yourself writing a calculated column just to total it up later, that is the sign it should have been a measure.
This is exactly where Power BI gets tricky Knowing whether to write a measure or a calculated column is one of the first real judgement calls in Power BI, and the concept of context underneath it is where most learners stall. Our two-day, hands-on Power BI Masterclass teaches both the practical way, on real models. For a wider view of the options, our guide to choosing Power BI training in the UK explains what to look for in a course.
