Product Sales
Database Design - Data Manipulation
Task-based
Overview
The purpose of this project is to use the spreadsheet sales data of a company and build a database from it. In this case, the company uses only spreadsheets to maintain their data (no formal database). Using their spreadsheet data, I will run them into Microsoft SQL Server and build a database around it creating tables, defining relationships, and making views that can be used by the employees to analyze the company's retail sales and support customer service operations.The second part of this project will be to give an example of the type of analysis that can be conducted once the database has been completed. This includes providing insights into what states had the highest sales, finding the months with greatest percentage increase in sales from the prior month, and products commonly sold together.
Project inspiration and original dataset found here.
If PDF does not render below it can be found on my Github here. Full Product Sales (Database Design/Analysis) project code can be found on my Github here.
Product Sales - PDF
Analysis Points
- California ($13.7M), New York ($4.7M), and Texas ($4.6M) were the top 3 states for sales for 2019 (their annual sales totals were all greater than the entire market's average sales - $4.3M).
- October had the highest percentage increase in sales from the previous month at 78.32%, followed by December at 44.10%. This shows that customers start the majority their holiday shopping in October, despite Black Friday deals in November where sales percentages dropped 14.38%.
- The top 2 paired product groupings were 1) Lightning Charging Cable/iPhone and 2) Google Phone/USB-C Charging Cable (not much insight can be drawn from this since most smartphones come with cables (not necessarily chargers) included - it could be determined that customers are just purchasing extra cables for "just in case".