This is my personal project to illustrate my Business Intelligence skills using Microsoft Power BI.
New Market is a fictional multi-national supermarket company having stores in Canada, Mexico, and the United States. The raw data files were provided in .csv format. (Data Source: Maven Analytics)
I have worked through the entire business intelligence workflow – connecting and shaping the source data, building a relational model, adding calculated columns and measures, and designing an interactive report.
Connecting & shaping the data
Imported all the raw data .csv files into Microsoft Power BI project. Followed proper naming conventions and promoted headers for each file. Also confirmed that data types are accurate. I used merge, extract, replace, conditional columns, statistical tools, and data tools for data shaping.
Creating the data model
Created this Data Model in the relationships view, and arranged dimension (lookup) tables above fact (data) tables for a clean and tidy look. I have used the “snowflake” schema to connect Stores to Regions.
Additionally confirmed that
– all relationships follow one-to-many cardinality, with primary keys (1) on the dim side and foreign keys (*) on the fact side.
– Filters are all one-way
-Filter context flows “downstream” from dim tables to fact tables.
Adding calculated columns and DAX measures
Added multiple calculated columns in the data view and DAX measures in the report view. This enabled me to extract and optimize critical data from raw data.
Building the report
This is the final phase of the project. I used a variety of matrix visuals available in Microsoft Power BI to create an easy-to-understand Annual Sales report for 1998. Added report level filter to only show data for 1998.
