Hi eDNA Members and Staff,
First of all thank you for this opportunity to develop our skills and please find below my submission for this 2nd week of the PBI Accelerator.
Here are some comments about the work I have done.
When creating the dimensions tables especially the “Products” table, the “Customers” table and the “Locations” table, I first investigated the data to checked if there are only unique values or some multiple values.
-
Customers table
The couple (Customer ID , Customer Name) did not revealed any multiple values. -
Locations table
One Postal Code “92024” have two different cities “Encinitas” and “San Diego”. After checking on “Postal Codes United States (geonames.org)”, I replaced “San Diego” by “Encinitas” where Postal Code = “92024”
Some Postal Codes were not matching the correct cities so I made adjustments but I did not went through each Postal Code. -
Products table
As mentioned before, 32 Products ID have two Product Names. I adjusted the Product ID of each one of the two Product Name that have the same Product ID by adding a suffix to the Product ID.
I made the adjustments in the “Client Flat File” query and I left the queries I used to check for multiples
For the “Date Table” I used “Melissa de Korte” script provided on eDNA Extended Date Table (Power Query M function)
I also applied these steps
- Marking the Dates table as Date Table
- Defining the Summarizing By to None for all columns
- Categorizing the columns of the Location tables
- Sorting by Column
- Hiding columns used for sorting other columns
- Creating some basic measures.
EDNA Power BI Accelerator – Week 2 Problem Final - MehdiH.pbix (10.6 MB)
Any comments will be appreciated.
Best Regards
Mehdi HAMMADI