Exploratory Analysis for Walmart Sales Data

Overview

This project focuses on analyzing Walmart sales data to identify top-performing branches and products, as well as understanding sales trends and customer behavior. The goal is to uncover insights that can help optimize sales strategies. The sales dataset was a modified version of the dataset from the Kaggle Walmart Sales Forecasting Competition. The modified dataset consisted of sales transactions from six product lines in three Walmart branches. All sales transactions occurred in October, November, and December 2023.

Columns

Column	Description	Data Type
invoice_id	Invoice ID for the sales made	VARCHAR(15)
branch	Branch where the sales occurred	VARCHAR(5)
city	Location of the branch	VARCHAR(15)
customer_type	Type of customer	VARCHAR(10)
gender	Gender of the purchasing customer	VARCHAR(10)
product_line	Product line of the purchased item	VARCHAR(50)
unit_price	Price per product	DECIMAL(10, 2)
quantity	Quantity of product sold	INT
tax_pct	Tax on the purchase	FLOAT(6, 4)
total	Total cost of the purchase	DECIMAL(12, 4)
date	Date of the purchase	DATE
time	Time of the purchase	TIMESTAMP
payment	Method of payment	VARCHAR(15)
cogs	Cost of Goods Sold	DECIMAL(10, 2)
gross_margin_pct	Percentage of gross margin	FLOAT(11, 9)
gross_income	Gross income generated	DECIMAL(12, 4)
rating	Customer rating	FLOAT(2, 1)

Methods

1. Data Wrangling

Perform an initial inspection of the data to detect missing or NULL values. Replace or remove NULL values during database creation by enforcing NOT NULL constraints.

2. Feature Engineering

time_of_day: Categorize sales into Morning, Afternoon, and Evening to identify peak sales periods. day_name: Extract the day of the week (Mon, Tue, Wed, etc.) to determine when each branch is busiest. month_name: Extract the month (Jan, Feb, Mar, etc.) to analyze which months have the highest sales and profits.

3. Exploratory Data Analysis (EDA)

Exploratory data analysis is conducted to answer key questions and meet the objectives of the project, providing deeper insights into sales performance and trends.

Insights

1. Product Analysis

The Electronics & Clothing and Apparel are the top-performing product lines.

Electronics generated the highest total sales at $62,889.78 with a gross income of $4,658.50 and an average customer rating of 7.11. It also ranks highest in quantity sold (952 units). Clothing and Apparel ranked second in generating the highest total sales of $60,728.29 and a gross income of $4,498.39. It also maintains a strong customer rating of 7.03.

The Health and Wellness product line needs improvement.

While it generated a total sales of $54,807.82, there are rooms for major improvement in gross income and total sales.

2. Sales Analysis

Peak Sales Time

Most sales occured in the evening. This suggests that evening hours serve as a critical sales window, and sales strategies could be optimized around this peak period, such as increasing marketing efforts or staffing during these hours.

Busy Days by Branch

I found different trends across different branches on weekdays:

• West Branch: The busiest day of the week is Friday.
• East Branch: Thursday marks the busiest sales day.
• South Branch: Both Sunday and Thursday are noted as peak sales days.

These insights indicate that targeted promotions, inventory consideration, or special events could be more effectively aligned with these specific days to capitalize on peak sales periods.

Peak Profit Month

The month of December generated the most profit across all three branches. This peak profit is highly likely a result of the holiday season to celebrate Christmas and New Year. Walmart should maintain robust marketing campaigns and enough stock levels for the month of December.

3. Customer Analysis

Customer Type

​Customers with memberships at Walmart tend to buy slightly more often than their non-member counterparts.​ This trend suggests that membership programs may enhance customer loyalty and encourage higher purchasing frequency, making membership a potentially profitable avenue to explore further.

Peak Rating Time

Examination of rating data reveals that the most ratings are provided by customers during the evening hours. This pattern implies that engagement, possibly due to more available time after work, allows customers to provide feedback on their shopping experiences. Businesses could leverage this insight by timing review requests to coincide with this peak period, thereby increasing the likelihood of receiving responses.

Best Days for Ratings

On analyzing average ratings by day of the week, Wednesdays, Saturdays, and Sundays emerged as the days with the best average ratings. This indicates customer satisfaction may peak during mid-week and the weekend, possibly due to promotional activities, special events, or an overall more relaxed shopping environment on these days. Walmart should consider running promotions or marketing campaigns strategically on these days to maximize customer interaction and satisfaction.