The ArrayAggDistinct function identifies distinct non-null row values in a column or group and aggregates them into a single array.


This function is supported by AlloyDB, BigQuery, Databricks, PostgreSQL and Snowflake connections.


ArrayAggDistinct(value, [direction])

Function arguments:

value(required) The column containing values to join.
direction(optional) The directional order to sort the column values.

  • “asc” sorts values in ascending order.
  • “desc” sorts values in descending order.

When a direction isn’t specified, Sigma sorts values in the same order as the column referenced in the value argument. If the referenced column isn’t sorted, Sigma sorts values in ascending order by default.


ArrayAggDistinct([Product Type], “asc”)

For each grouping (grouped by order number) the ArrayAggDistinct function returns an array containing distinct non-null values—sorted in ascending order—from the corresponding rows in the Product Type column.

Table showing product data grouped by order number, with a column aggregating distinct product type values using the ArrayAggDistinct function.