Group Rows By Multiple Columns

Use more than one column in `GROUP BY` when each combination defines a separate aggregate bucket.

4/4 supported engines validation-green 1 example 2 scenarios

Docker-validated Not currently validation-green

Example 1

Count tickets by team and status

Each unique pair of team and status becomes its own result row. Billing has one open ticket and two closed tickets; Support has two open tickets and one closed ticket.

Source table data Rows loaded before the example query runs.

Setup

CREATE TABLE tickets (id INT, team VARCHAR(20), status VARCHAR(20));

INSERT INTO
  tickets (id, team, status)
VALUES
  (1, 'Support', 'open'),
  (2, 'Support', 'open'),
  (3, 'Support', 'closed'),
  (4, 'Billing', 'open'),
  (5, 'Billing', 'closed'),
  (6, 'Billing', 'closed');

Validated query Shared across supported engines.

SQL

SELECT
  team,
  status,
  COUNT(*) AS total_tickets
FROM
  tickets
GROUP BY
  team,
  status
ORDER BY
  team,
  status;

Expected result Returned rows for the shared example.

team	status	total_tickets
Billing	closed	2
Billing	open	1
Support	closed	1
Support	open	2

All engines group by the same `(team, status)` combinations and return the same counts here.

Useful when

Where this command helps.

counting records per status within each region or team
building reports where the aggregate bucket depends on more than one dimension

Explanation

What the command is doing.

Grouping by multiple columns creates one aggregate row per unique combination of those values. This is useful when a single grouping column is too broad, such as when you need counts by both region and status, or totals by both month and category.

Group Rows By Multiple Columns

Count tickets by team and status

Where this command helps.

What the command is doing.

Keep moving through nearby tasks.