Retail Store Acquisition Analysis with Python

View project code here: Maven Mega Mart Analysis

This project is a modified version of the midcourse project from the Maven course: Data Analysis with Python and Pandas.

Maven Mega Mart is considering acquiring a new chain of stores. Data was analyzed to determine the distribution of sales across customers, products, and stores.

To determine suitability for acquisition Maven needs to learn; 1) Which products are generating the most sales, 2) who are the customers and how are sales distributed across the customer base, and 3) which stores are generating the most revenue and which are underperforming.

This project aims to provide insights rather than a definitive recommendation on the acquisition.

Below are the steps followed to analyze the data:

1. Import the necessary libraries and the data

2. Explore the data.

• View the first and last rows of the data, plus a random sample
• Count the rows and columns. View the data types and memory usage
• Took a statistical snapshot of the data
• Checked for nulls and unique values

3. Transform the data.

• Added and dropped columns as necessary

4. Analyze the data with a focus on the following;

• Total sales, Total discounts, and Quantity sold
• Average sales value per basket, Average sales value per household, and Sales comparison of individual products
• Number of stores, top stores by sales, bottom stores by sales
• Product comparison by sales value
• Household comparison by sales value

5. Visualize the data

• Box plots, column charts, bar charts, and histograms were used during analysis to visualize the data

6. Conclusions

• There are many repeat customers which may indicate loyalty or a lack of alternatives.
• These stores gain much of their revenue from staple items like milk, bread, and gasoline. This has value since staples often form a defensive product line during economic downturns.
• Staple products sold at these stores are not as reliant on discounts as other products which are sold with higher discount rates.
• Many of the top products at these stores are perishable items. Consistent turnover of these products is necessary to avoid losses. This may at times require discounting but still appears to be less than the average discount.
• The average price per product is under 5 dollars, so high volume is necessary to achieve high revenue. Analysis of the profit margin per product would provide additional insight.
• The majority of sales revenue appears to come from a small number of stores. More information is needed about non-performing stores.
• Gasoline is one of the top-selling products at these stores. This may work well now but performance projections more than 5 years out may need additional consideration of changing consumer preferences for electric vehicles.
• The data only included information about store revenue. To form a more complete picture of how the stores perform, cost would also need to be analyzed.