A measure is a DAX formula that Power BI calculates when a visual asks for it, using the filters active in that visual. A calculated column is a DAX formula that Power BI calculates once per row when the data refreshes, and stores in the table. Use a measure for totals, ratios and KPIs that should change with slicers, and a calculated column for a per-row value you need to slice, filter, group or join on.

If a calculation works in one visual and not another, the choice between the two is often the reason.

What Is a Measure in Power BI?

A measure is evaluated at query time, in the filter context of the cell it appears in. The same measure returns a different number in every row of a table visual, because each row filters the data differently. It reacts to:

• slicers

• filters on the visual, page and report

• the rows, columns and axis of the visual itself

• cross-filtering from other visuals

A measure is not stored in the model. Only its formula is.

Example of a Measure

Total Sales = SUM ( Sales[Revenue] )

In a table visual split by product category, this measure returns sales for each category. Filter the report to 2025 and it recalculates for 2025. Measures are the right place for totals, averages, ratios and KPIs. A measures table keeps them organized.

What Is a Calculated Column in Power BI?

A calculated column adds a new column to an existing table. Power BI evaluates its formula once for every row, when the table refreshes, and stores the results in the model like any imported column.

Example of a Calculated Column

Profit = Sales[Revenue] - Sales[Cost]

This stores a profit value on every row of the Sales table. Slicers do not change the stored values. A slicer only decides which rows are visible, and a measure that sums the column then adds up those rows.

Calculated columns are best for row-level logic: categories, flags, keys for relationships and values you want on an axis.

Measures vs Calculated Columns: Key Differences

The Biggest Difference: Dynamic vs Static Calculations

• Measures are dynamic. Filter by region and the measure recalculates for that region.

• Calculated columns are static. The values stay the same until the next refresh. If the semantic model refreshes daily, the column changes daily.

When Should You Use a Measure in Power BI?

Use a measure when the result must change with the report context.

Common Use Cases for Measures

• Total sales and total customers

• Year-to-date and other time intelligence

• Profit margin

• Average order value

• Conversion rate

• Growth percentages

Example: Profit Margin Measure

Profit Margin % =
DIVIDE (
    SUM ( Sales[Revenue] ) - SUM ( Sales[Cost] ),
    SUM ( Sales[Revenue] )
)

This adjusts to region, product, date or any other filter. A margin must be a measure. A calculated column holds each row's own margin, and adding up row margins does not give the margin of the total.

Why Measures Are Usually the Best Option

• They respond to every filter, so one measure serves every visual.

• They add nothing to model size.

• They support time intelligence, ratios and conditional logic across rows.

When Should You Use a Calculated Column in Power BI?

Use a calculated column when you need a per-row value that becomes part of the table.

Common Use Cases for Calculated Columns

• Categories such as High, Medium and Low

• Yes/No flags

• Year or month parts of a date, when there is no date table

• Keys for relationships

• Custom groupings

Example: Sales Category Column

Sales Category = IF ( Sales[Revenue] > 1000, "High", "Low" )

This column can go in a slicer, a filter or on the axis of a chart.

Can Measures Be Used as Filters or Slicers?

Not in a slicer and not in a relationship. Both need stored column values. A measure can be used as a visual-level filter in the Filters pane, for example "show items where Total Sales is greater than 1,000". The Power BI filters guide covers the filter types.

Performance differences between Measures vs Calculated Columns

• Measures cost CPU at query time. A simple measure over a well-designed model is fast. A complex one can slow down every visual that uses it.

• Calculated columns cost memory and refresh time. Every row stores a value, and every refresh recalculates the whole column. The effect grows with the number of rows and the number of distinct values in the column.

The rule of thumb: if the result is an aggregation, write a measure. If it is a per-row attribute, a column is fine, and it is often better built in Power Query or the source system than in DAX. The DAX performance guide covers how to find which one is slow.

Understanding Row Context vs Filter Context

This is why the two behave differently.

• Calculated columns run in row context. The formula sees one row at a time, so Sales[Revenue] - Sales[Cost] means "this row's revenue minus this row's cost".

• Measures run in filter context. The formula sees the set of rows that the visual, slicers and filters leave visible.

A calculated column that aggregates needs CALCULATE. In a Customer table, SUM ( Sales[Revenue] ) returns the revenue of all customers on every row, because row context does not filter the Sales table. CALCULATE ( SUM ( Sales[Revenue] ) ) turns the current row into a filter and returns that customer's revenue. This is called context transition. The row context vs filter context post explains it in full.

Aggregation and Iterator Functions in DAX (SUM vs SUMX Explained)

Aggregators such as SUM, AVERAGE, MIN, MAX and COUNT work on one column. Iterators such as SUMX and AVERAGEX calculate an expression row by row inside a measure, then aggregate. So a measure like SUMX ( Sales, Sales[Revenue] - Sales[Cost] ) gives total profit without storing a Profit column. The SUM vs SUMX post explains when the two give different results.

Measures vs Calculated Columns: Best Practice Recommendations

Use a measure when:

• You need totals, averages, ratios or KPIs.

• The result must respond to slicers.

• You are doing time intelligence (YTD, MTD, rolling averages).

Use a calculated column when:

• You need a value per row.

• You need the value in a slicer, on an axis or for grouping.

• You need a relationship key.

• You need segmentation labels.

Common Mistakes People Make

1. Too many calculated columns. Each one adds memory and refresh time. Build per-row columns in Power Query or the source where you can.

2. Trying to put a measure in a slicer. Slicers need a column. If readers must pick a band such as High or Low, build the band as a column or a separate table.

3. Business KPIs as calculated columns. A margin or conversion rate stored per row cannot be added up correctly. KPIs belong in measures.

4. Aggregations in Power Query. Power Query is the right place to shape and clean rows. Totals that must follow the report filters belong in measures.

Measures vs Calculated Columns: Quick Decision Cheat Sheet

FAQs

Are measures better than calculated columns in Power BI?

For totals, ratios and KPIs, yes. Measures follow every filter and add nothing to model size. For per-row labels, flags and keys, a calculated column is the right tool.

Can I use a measure in a slicer?

No. A slicer needs a column with stored values. You can use a measure as a visual-level filter in the Filters pane.

Do calculated columns slow down Power BI?

They can. Each one adds memory to the model and time to every refresh, and the effect grows with row count and the number of distinct values. They do not usually slow down visuals that do not use them.

When should I use calculated columns instead of measures?

When you need a per-row value in a slicer, on an axis, in a relationship or as a grouping label.

Sources

• Using calculated columns in Power BI Desktop - Microsoft Learn

• Measures in Power BI Desktop - Microsoft Learn

• CALCULATE function (DAX) - Microsoft Learn

• SUMX function (DAX) - Microsoft Learn