Joining 3 tables in SQL showing all 3 tables - sql

Question

I am a basic user of SQL but need to join 3 tables together to sow a) sales b) returns and c) profit
I currently have the following code
select * from (
select SUM(Return_Amount) , 'Return' as type, monthname(Return_Date) as month_
from returns
group by month_
union
select SUM(Order_Total_Cost) , 'Sales' as type, monthname(Order_Date) as month_
from sales
group by month_
union
select SUM(profit) as profit_ , 'Profit' as type, month_
from(
select sell_price-cost_price as profit , monthname(order_date) month_
from sales
join order_item
on order_item.order_No = sales.order_No
join returns
on returns.order_no = sales.order_No
join supplier
on supplier.Product_ID = order_item.Product_ID
) B group by month_
) A order by month_;
This is showing as below:
387 Return August
182 Sales August
867 Profit August
733 Return July
109 Sales July
646 Profit July
596 Return June
I want it to show with Return, Sales & Profit as separate columns instead of all types listed in one.
Any help would be greatly appreciated.
Thanks

Answer 1

Why not join them?
SELECT x.month_
, x.returns_
, y.sales_
, z.profit_
FROM
(SELECT SUM(Return_Amount) AS returns_
, monthname(Return_Date) AS month_
FROM RETURNS
GROUP BY month_) x
INNER JOIN
(SELECT SUM(Order_Total_Cost) AS sales_
, monthname(Order_Date) AS month_
FROM sales
GROUP BY month_) y ON x.month_ = y.month_
INNER JOIN
(SELECT SUM(profit) AS profit_
, month_
FROM
(SELECT sell_price-cost_price AS profit
, monthname(order_date) month_
FROM sales
INNER JOIN order_item ON order_item.order_No = sales.order_No
INNER JOIN RETURNS ON returns.order_no = sales.order_No
INNER JOIN supplier ON supplier.Product_ID = order_item.Product_ID) B
GROUP BY month_) z ON x.month_ = z.month_

Answer 2

You could turn the three unioned queries to subqueries and join them. This assumes that each subquery always produces one record per month.
Also: you probably want to join and month and year, in case your data spans over multiple years (which eventually happens in any live dataset). I would also recommend using numeric years and month (as returned by year() and month()), which would produce more efficient joins than month names.
select
r.month_name,
r.total_return_amount,
s.total_sales,
p.total_profit
from
(
select
year(return_date) yr,
month(return_date) mh,
monthname(return_date) month_name
sum(return_amount) total_return_amount,
from returns
group by
year(return_date),
month(return_date),
monthname(return_date)
) r
inner join (
select
year(order_date) yr,
month(order_date) mh,
sum(order_total_cost) total_sales
from sales
group by
year(order_date),
month(order_date)
) s on s.yr = r.yr and s.mh = r.mh
inner join (
select
year(order_date) yr,
month(order_date) mh,
sum(sell_price - cost_price) as total_profit ,
from sales
inner join order_item on order_item.order_no = sales.order_no
inner join returns on returns.order_no = sales.order_no
inner join supplier on supplier.product_id = order_item.product_id
group by
year(order_date),
month(order_date)
) p on p.yr = r.yr and p.mh = r.mh