similar SKUs flow - SAC

[Deb_]

I'm facing an SAC challenge.
I have a dimension of SKUs with various properties such as weight, use, dosage, market, and package.
Each SKU has an associated cost uploaded through ETL.

My objective is to validate the accuracy of the uploaded costs.
To achieve this, I plan to create a systematic flow that analyzes the reasonability of the costs.
The approach involves grouping SKUs with a specific number of common attributes and then applying a threshold to identify significant cost differences.
For instance, I will start by grouping SKUs with 7 common attributes and examine cost differentials.
Subsequently, I will repeat the process for SKUs with 6 common attributes, and so on.

Can you guide me on how to construct an effective flow for this cost validation process?

thank you

[N1kh1l · Accepted solution]

@Deb_ 

If you have 7 attributes like weight, use, dosage, market, and package etc. You will need 6 more attributes or may be more if you want to create different combinations of attributes for grouping and this will be combination of this attributes. You can just concatenate all attribute values as it is or come up with some coding sequence for attribute values and then concatenate them together. So it can be like Weight+Use+dose+Market+package , Weight+Use+dose+Market , Weight+Use+dose , Weight+Use etc. You can create more combination based on your analysis requirement. Once you have these attributes filled in the ETL load. Just create a table using this attributes rather then members and you will get all aggregation based on this grouping which can help you see cost of SKU based on attribute groupings.

Br.

Nikhil