SQL query needed for best product mix data - sql

Question

I work for a small company dealing with herbal ingredients. We count regularly the effectiveness of the ingredients, based on the "product mix" (how much of the ingredient A, B and C). I have a table with thousands of rows, like the following:
PRODUCT Ingredient A Ingredient B Ingredient C EFFECTIVENESS
1 A                          28                               94                            550                      4,1
2 B                         50                               105                           400                      4,3
3 C                         30                               104                           312                       3,5
.. Etc etc etc etc Etc
What   I want as a result, is the table below. I am using excel during the last years however it is difficult to handle millions of data and therefore I would like now to have something similar in sql. I did several attempts with Pivot and subqueries but I did not manage to get the result I needed.
In particular, in the first three columns, I include various ranges / criteria. In the column ‘Average effectiveness’ it is counted the average effectiveness of the ‘total products’ which meet these criteria. Due to the fact that the ranges are hundreds e.g. for Ingredient A, I have more than 100 different ranges and similarly for Ingredient B and C, I would like a way to have all multiple combinations of A, B, C ingredients (ranges) automatically.
Ingr. A    Ingr. B     Ingr. C      Total products    Average Effectiveness
1-10       50-60       90-110      ???                            ???
1-10       50-60      110-130     ???                             ??
1-10       50-60      130-150     ????                           ??
1-10       60-70      150-170     ???                             ??
10-20     60-70      90-110       ???                             ??
10-20     60-70      110-130      ???                            ??
10-20     60-70      130-150      ??                              ??
Etc etc

Answer 1

I'm unable to give a more specific answer, but I think what you need to do is;
Use the CUBE to get all of the combinations and to aggregate the SUM and AVG values
Summarizing Data Using CUBE
The CUBE query will take its data from a nested query that has your data stored by the range of a value rather than the actual value. You can refer to SQL's CASE expression for more information on transforming the data so that it stores the range of a value rather than the value.
So, in other words, first you transform your data so that you're storing which range a value occurs in. Then from that transformed data, you summarize it using the CUBE to get all the combinations. So #1 is the outer query and #2 is the inner query.
Here is a very rough idea of what the query might look like, just to give you an idea:
Select Ingr_A, Ingr_B, Ingr_C, COUNT(*), AVG(Effectiveness)
(SELECT
Product,
Effectiveness,
"Ingr_A" =
CASE
WHEN Ingredient_A >= 10 and Ingredient_A < 20 THEN '[10, 20)'
WHEN Ingredient_A >= 20 and Ingredient_A < 30 THEN '[20, 30)'
...
END,
"Ingr_B" =
CASE
(like above)
END,
"Ingr_C"
(etc.)
FROM ProductsTable)
GROUP BY Ingr_A, Ingr_B, Ingr_C WITH CUBE