SUMMARIZE

A table's groups by columns of it or of related tables, optionally with columns computed for each group.

Group: Table manipulation. Compatible with DAX.

Syntax

SUMMARIZE ( <table>, <groupBy_columnName> [, <groupBy_columnName>] … [, <name>, <expression>] … )

Parameter	Type	Description
`table`	Table	The table whose rows are grouped, often a fact table such as Sales.
`groupBy_columnName`	Column repeatable	A column to group by: the table's own, or a column of a table it reaches through relationships, such as Products[Category] from Sales.
`name`	Text optional, repeatable	The name of an added column, in quotes.
`expression`	Expression optional, repeatable	Its value for each group, with the group as filter.

Returns

Table. One row for each combination of the grouping columns' values that occurs in the table's rows, with any added columns.

Remarks

Only combinations that occur come out: grouping Sales by category and kind of place gives no depot rows, as depots sell nothing. CROSSJOIN would list them.
Columns of related tables are reached along the relationships, from the many side to the one side. A column the table cannot reach is an error.
For added columns, prefer ADDCOLUMNS around SUMMARIZE, as in the second example: name and expression pairs inside SUMMARIZE work, but in a context that is harder to predict.
With ROLLUP among the grouping columns it adds subtotal rows, which ISSUBTOTAL tells apart.

Examples

The combinations that occur

EVALUATE
SUMMARIZE ( Sales, Products[Category], Places[Kind] )
ORDER BY Products[Category], Places[Kind]

Products[Category]	Places[Kind]
Beer	Shop
Cider	Shop

Groups with computed columns

EVALUATE
ADDCOLUMNS (
    SUMMARIZE ( Sales, Products[Category] ),
    "Liters", [Liters sold],
    "Shops", CALCULATE ( DISTINCTCOUNT ( Sales[Shop] ) )
)
ORDER BY Products[Category]

Products[Category]	Liters	Shops
Beer	1505389	15
Cider	301074	15

A table made in the query

EVALUATE
VAR Sold =
    DATATABLE (
        "Shop", STRING, "Product", STRING, "Liters", INTEGER,
        { { "A", "Pils", 10 }, { "A", "IPA", 5 }, { "B", "Pils", 7 }, { "B", "Pils", 3 } }
    )
RETURN
    SUMMARIZE ( Sold, [Shop] )
ORDER BY [Shop]

Shop
A
B

SUMMARIZECOLUMNS, ADDCOLUMNS, GROUPBY, ROLLUP, VALUES.

Last updated 3 November 2026 · OpenBI Studio 0.3.74