Question

I’m looking for a single MySQL query (subqueries are fine) that will get a distribution of units per order over the past year in the following format:

units_per_order     |     number_of_orders     |     percent_of_total
---------------------------------------------------------------------
1                   |                 7500     |                  55%
2 
3 
4
5
6-10
10-20
30-50
50–100
100+

EDIT:

Tables and necessary columns to query against:

Table: orders
Fields: orders_id, date_purchased

Table: orders_products
Fields: orders_id, products_quantity

This is what I have right now:

SELECT COUNT(*) AS cnt, qty FROM (
        SELECT SUM(products_quantity) as qty, orders_id
        FROM orders_products op LEFT JOIN orders o USING (orders_id)
        WHERE date_purchased LIKE '2012%'
        GROUP BY orders_id
    ) AS new_table WHERE qty != 0 GROUP BY qty ORDER BY qty

Any ideas?

Answer 1

Ok, here’s what I ended up with. Please advise if you see any way to make it more efficient.

SELECT qty, cnt, ROUND(((cnt / total) * 100)) AS percent FROM (
    SELECT qty, COUNT(*) AS cnt, total
    FROM (
        SELECT
            (
                CASE WHEN (SUM(products_quantity) < 6) THEN SUM(products_quantity)
                    WHEN (SUM(products_quantity) BETWEEN 6 AND 10) THEN 6
                    WHEN (SUM(products_quantity) BETWEEN 11 AND 20) THEN 11
                    WHEN (SUM(products_quantity) BETWEEN 21 AND 50) THEN 21
                    WHEN (SUM(products_quantity) BETWEEN 51 AND 100) THEN 51
                    ELSE 101
                END
            ) AS qty,
            orders_id,
            (
                SELECT COUNT(DISTINCT(orders_id))
                FROM orders JOIN orders_products USING (orders_id)
                WHERE date_purchased LIKE '2012%' AND products_quantity != 0
            ) AS total
        FROM orders_products op JOIN orders o USING (orders_id)
        WHERE date_purchased LIKE '2012%'
        GROUP BY orders_id
    ) AS new_table
    WHERE qty != 0 GROUP BY qty
) AS newer_table