Sort Rows

Order rows by one or more columns, each with its own ascending or descending direction.

The Sort Rows step reorders the rows of the dataset by the columns you choose.

Configuration

Add one or more sort columns. Each one carries its own direction, so you can sort by date descending and then by country ascending in the same step.

Field	Description
Column	A column to sort by. Columns apply in the order you add them: the first is the primary sort key, later ones break ties.
Sort order	Direction for that column — ascending (lowest to highest, A to Z) or descending.

At least one sort column is required; the step fails if none is selected, or if a selected column is not present in the data at that point in the pipeline.

Sorting changes only row order — no rows are added, removed or recalculated.

Sorting matters when a later step depends on row order, such as Data snapshot or Publish dataset. It does not affect group or rolling-window functions in Calculate Variable, which resolve their own ordering by date and entity.

To place rows in an explicit, arbitrary sequence rather than sorting by value, use the Reorder Rows step instead.

Next steps

Filter rows

Drop rows that don’t meet a condition.

Reorder columns

Change column order rather than row order.