Problem Statement:
An online retail store is trying to understand the various customer purchase patterns for their firm, you are required to give enough evidence based insights to provide the same.
Dataset Information:
The online_retail.csv contains 541909 rows and 8 columns.
- Invoice - Invoice number
- StockCode - Product ID
- Description - Product Description
- Quantity - Quantity of the product
- InvoiceDate - Date of the invoice
- Price - Price of the product per unit
- CustomerID - Customer ID
- Country - Region of Purchase
Problem Objective:
- Using the above data, find useful insights about the customer purchasing history that can be an added advantage for the online retailer.
- Segment the customers based on their purchasing behavior.
Steps involved:
- Data Preprocessing and EDA
- Feature Engineering
- Data Visualization
- Correlation analysis
- RFM analysis
- Machine Learning Model Development
- Model Evaluation
Conclusion:
Strategic Actions Based on Segments:
- Best Customers (3054 Customers, $4,150,536.51 Total Sales):
Action:
- Focus on maintaining their loyalty.
- Provide personalized offers, loyalty rewards, and early access to new products.
- Regularly engage with these customers through targeted communication.
- Loyal Customers (1067 Customers, $512,818.85 Total Sales):
Action:
- Re-engage with these customers.
- Use email campaigns, special discounts, and personalized messages to remind them of your products.
- Offer incentives to bring them back.
- Big Spenders (13 Customers, $1,655,398.08 Total Sales):
Action:
- Treat them as VIPs.
- Provide top-tier loyalty programs, exclusive access to premium products, and personalized concierge services.
- Ensure excellent customer service and maintain close relationships.
- At Risk (204 Customers, $2,592,654.46 Total Sales):
Action:
- Implement win-back strategies.
- Send personalized offers and discounts, conduct satisfaction surveys, and offer incentives to retain them.
- Identify potential reasons for their decline in activity and address these issues.