Retail Sales Analysis
A US-based retail store has provided historical sales data, capturing
transactions across different product categories, customer segments, and regions. The
company aims to analyse store performance, customer purchasing behaviour, and the
impact of discounts on profitability.
The goal is to:
Segment customers based on purchasing behaviour and loyalty.
Identify sales trends across different time periods and regions.
Evaluate discount strategies to optimize profitability.
Provide AI-driven business recommendations to improve sales, marketing
effectiveness, and customer retention strategies.
Dataset: [Link]
Data Description:
Order ID A unique identifier for each order placed by a customer.
The date when the order was placed, useful for time-series
Order Date
analysis and sales trend predictions.
The date when the order was shipped, useful for analyzing
Ship Date
shipping times and operational efficiency.
Customer A unique identifier for each customer, allowing for customer
ID segmentation and lifetime value analysis.
Customer The name of the customer, often used for customer-centric
Name reporting.
The business segment of the customer, such as Consumer,
Segment
Corporate, or Home Office.
The geographical region where the sale occurred (e.g., East, West,
Region
Central, South).
Product The category of the purchased product, such as Furniture, Office
Category Supplies, or Technology.
Product
A more specific classification of the product (e.g., Chairs, Phones,
Sub-
Binders).
Category
Product ID A unique identifier for each product in the inventory.
Product
The name or description of the product.
Name
The number of units sold for a particular product in a single
Quantity
transaction.
The discount applied to the product during the transaction,
Discount
expressed as a percentage.
Sales The revenue generated from the transaction (after discounts).
Profit The monetary profit earned from the transaction (Sales - Cost).
Shipping The method of shipping used (e.g., Standard Class, Second Class,
Mode Same Day).
Initial Guidelines:
Ensure to follow to Use Id’s provided by UNext for naming file as
conventions.
Task: Load the dataset and perform preliminary EDA (Exploratory Data
Analysis) with key observations and insights- (weightage - 50 marks)
T1: Load the store dataset (Weightage – 4 marks)
Load the dataset using Pandas. (Hint: use encoding=’latin’ argument)
Display the first 5 rows.
T2: Data Cleaning (Weightage – 3 marks)
Count duplicate rows and remove them.
T3: Uniqueness Check (Weightage – 3 marks)
Find unique values in "Customer Segment", "Ship Mode", "Product
Category"
T4: Overall Sales Performance (Weightage – 8 marks)
What is the total sales volume (Total Amount) over the entire dataset
period?
Are there any missing values in the data? If yes, handle the missing
values by using appropriate technique.
T5. Customer Segmentation (Weightage – 6 marks)
Which customer segment (Consumer, Corporate, Home Office)
contributes the most to total sales and profits?
Which regions have the highest number of high-spending customers?
T6: Product Category Analysis (Weightage – 5 marks)
Which Product Categories are most popular in terms of quantity sold and
total sales revenue?
T7: Calculate new metrics (Weightage – 6 marks)
Shipping time: Ship Date - Order Date to get delivery time.
Discounted Price: Sales - (Sales * Discount) to determine the actual price
after applying the discount.
T8: Identify Most Discounted Products (Weightage – 3 marks)
Find the top 10 products with the highest discount applied.
T9: Visualization (Weightage – 6 marks)
Create a correlation matrix and visualize it using a heatmap
Plot a boxplot to visualize outliers in the data.
T10: Data Transformation (Weightage – 6 marks)
Remove the unnecessary variable for model building process
[“Order_ID", “CustomerID”].
Apply normalization technique (standard scaler).
Apply encoding technique to convert categorical variable into numerical.
-----------------------------------------------------------------------------------------------