Build an activity table by signup cohort
Your task
Return one row per customer signup month. Show signup_month in YYYY-MM format, the distinct number of customers with completed orders in January 2025 as jan_buyers, and the distinct number in February 2025 as feb_buyers. Include cohorts with zero buyers in both months. Sort by signup month.
A signup cohort groups customers by their signup month. A customer who places several orders in one month is still one active buyer for that month.
Sample data
| id | name | signup_date | source |
|---|---|---|---|
| 1 | Avery | 2025-01-02 | organic |
| 2 | Casey | 2025-01-05 | |
| 3 | Jordan | 2025-01-12 | organic |
| 4 | Morgan | 2025-02-01 | |
| 5 | Riley | 2025-02-10 | social |
| 6 | Sam | 2025-02-20 | organic |
| 7 | Taylor | 2025-03-01 | social |
| 8 | Devon | 2025-02-07 |
Expected output
| signup_month | jan_buyers | feb_buyers |
|---|---|---|
| 2025-01 | 2 | 1 |
| 2025-02 | 0 | 2 |
| 2025-03 | 0 | 0 |
Show a hint
Start with customers, use a LEFT JOIN for completed orders, and use COUNT(DISTINCT CASE WHEN ... THEN c.id END) once for each activity month.
Customer and sales analytics guide
Define which orders count
This part of the record shop database tracks customer orders separately from the earlier album sales table. A completed order contributes its full price in cents. Refunded orders are fully refunded, and cancelled orders contribute zero. Decide whether you need to keep all statuses for a report or count only completed orders.
SELECT status, COUNT(*) AS orders
FROM shop_orders
GROUP BY status
ORDER BY status;Group at the customer level first
A customer can have several orders. Summarize by customer ID before counting customers or averaging customer-level values. To include customers with no completed orders, start with customers and put the completion filter in the LEFT JOIN condition.
SELECT c.id, COUNT(o.id) AS completed_orders
FROM customers AS c
LEFT JOIN shop_orders AS o
ON o.customer_id = c.id AND o.status = 'completed'
GROUP BY c.id
ORDER BY c.id;Choose a date boundary
Order and signup dates are stored as YYYY-MM-DD. A half-open date range includes its start and excludes its end. The example counts January orders by including January 1 and stopping before February 1.
SELECT COUNT(*) AS january_orders
FROM shop_orders
WHERE ordered_on >= '2025-01-01' AND ordered_on < '2025-02-01';Worked solution
SELECT STRFTIME('%Y-%m', c.signup_date) AS signup_month,
COUNT(DISTINCT CASE WHEN o.ordered_on >= '2025-01-01' AND o.ordered_on < '2025-02-01' THEN c.id END) AS jan_buyers,
COUNT(DISTINCT CASE WHEN o.ordered_on >= '2025-02-01' AND o.ordered_on < '2025-03-01' THEN c.id END) AS feb_buyers
FROM customers AS c
LEFT JOIN shop_orders AS o ON o.customer_id = c.id AND o.status = 'completed'
GROUP BY signup_month
ORDER BY signup_month;Press Control+Space for SQL suggestions. Use the arrow keys to select an option and Enter to insert it.
Results
Loading SQLite…
