Active in Every Quarter — Solution & ExplanationProblems / Active in Every Quarter / EditorialApproach

Reduce each order to its quarter. DATE_TRUNC('quarter', order_date) — four possible values across 2024.

Count both ways per customer. COUNT(DISTINCT quarter) measures *coverage*; COUNT(*) measures *volume*. The HAVING clause filters on coverage = 4; the SELECT reports volume.

Order by customer.

Key InsightThis is relational division — "rows related to *every* member of a set" — the query shape behind "watched all episodes", "passed every audit", "stocked in all regions". The elegant trick is turning a universally-quantified condition (for all quarters: an order exists) into a counting condition (distinct quarters = size of the required set), which SQL's aggregate machinery handles in one pass. Two details carry it: DISTINCT (volume in one quarter must not fake coverage) and knowing the divisor's size — hardcode 4 only when the required set truly is fixed.Hints

"Has orders in ALL four quarters" sounds like four EXISTS clauses — but it collapses to one aggregate condition: the customer's distinct quarter count equals 4.

COUNT(DISTINCT ...) is essential: four orders all in Q1 give COUNT(*) = 4 but only 1 distinct quarter. The data plants exactly that customer.

HAVING is where per-group conditions live, and it may reference aggregates that never appear in SELECT — filter on the distinct-quarter count, output the plain order count.

This works because quarters in 2024 can only take 4 values; when the universe of required values isn't fixed, compare against a subquery counting it instead of the literal 4.

Solve Active in Every Quarter yourself →