T-SQL Using CTE to aggregate totals for matching and non-matching periods - sql

Question

Matched Sales are provided by the join, It's getting the unmatched that is eluding me.
CTE
With PriorSalesCTE
(
Item
Variant,
Sum(sales)
Date between 7/1/2020 and 7/5/2020
),
CurrentSalesCTE
(
Item
Variant,
Sum(sales)
Date between 7/1/2021 and 7/5/2021
)
Select
SUM(cs.Sales) ‘MatchedSales’
FROM PriorSalesCTE ps join CurrentSalesCTE ps
ON cs.Item = ps.Item
And cs.Variant = ps.Variant
Now I need the empty spaces on both sides
I need the sales for items sold in 2020 but not sold in 2021 – Lost Sales
Conversely, sales for 2021 that did not sell in 2020 – New Sales.
I tried adding these in the CTE as separate sections of the CTE, but the join doesn’t give me what I need.
Any suggestions? Is the CTE simply preventing me for getting everything and maybe add a UNION ALL query to get the unmatched values?

Answer 1

For your actual query, you could use a FULL JOIN, which will give you the results from either side also.
But I think there is another solution: you don't need to join separate queries for this, you can just use conditional aggregation
WITH SalesByItem AS (
SELECT
t.Item,
t.Variant
Sales2020 = SUM(CASE WHEN Date BETWEEN '20200701' and '20200705' THEN t.Sales END),
Sales2021 = SUM(CASE WHEN Date BETWEEN '20210701' and '20210705' THEN t.Sales END)
FROM YourTable t
WHERE (Date BETWEEN '20200701' and '20200705'
OR Date BETWEEN '20210701' and '20210705')
GROUP BY
t.Item,
t.Variant
)
SELECT
NewSales = SUM(CASE WHEN Sales2020 IS NULL THEN Sales2021 END),
MatchedSales = SUM(CASE WHEN Sales2020 IS NOT NULL AND Sales2021 IS NOT NULL THEN Sales2021 END),
LostSales = SUM(CASE WHEN Sales2021 IS NULL THEN Sales2020 END)
FROM SalesByItem s;