Project Background

This project aims to segment customers of an online retail business to improve marketing effectiveness. The company operates in the e-commerce industry and has been active for several years. Its business model involves selling a variety of products to individual customers. A key business metric for the company is customer lifetime value, which is directly influenced by customer engagement, retention, and purchase frequency. This project provides insights and recommendations on the following key areas:

• Customer Segmentation: Identifying distinct customer groups based on purchase behavior.
• High-Value Customer Engagement: Strategies for retaining and maximizing revenue from top-tier customers.
• Marketing Optimization: Tailoring marketing efforts to specific customer segments.
• Outlier Customer Behavior: Understanding and addressing the behavior of unusual customer groups.

The full Python workflow, including data cleaning, RFM feature engineering, and K-Means clustering, is available in the Colab notebook.[Click here]

Data Structure & Initial Checks

The company's main database structure consists of a single table: the Online Retail dataset. The dataset contains transactional data. A description of the table is as follows:

• Online Retail Dataset: This table contains customer transaction data, with each row representing a single purchase of a product by a customer. Key columns include Invoice, StockCode, Description, Quantity, InvoiceDate, Price, Customer ID, and Country.

The data used for this project can be found here

Executive Summary

Overview of Findings

This project identified distinct customer segments based on their RFM (Recency, Frequency, Monetary) scores. The analysis revealed a significant variation in customer behavior, with a small segment of high-value customers contributing a disproportionate share of revenue. Targeted marketing strategies are recommended to maximize customer lifetime value across different segments.

Insights Deep Dive

Customer Segmentation:

• Main insight 1: Customers can be effectively segmented based on their RFM scores, revealing distinct groups with varying purchase patterns.
  • The analysis used K-Means clustering to identify four primary customer segments, in addition to outlier groups. These segments exhibit different levels of Recency, Frequency, and Monetary Value.
• Main insight 2: A significant portion of revenue is generated by a small group of high-value customers.
  • The "DELIGHT" segment, representing Monetary & Frequency Outliers, comprises customers with extreme spending and frequent purchases. These customers are the most valuable and require special attention.
• Main insight 3: Other segments show different engagement patterns, requiring tailored approaches.
  • The "PAMPER" segment consists of high spenders with infrequent purchases, while the "UPSELL" segment includes frequent buyers with lower spending per purchase.
• Main insight 4: RFM analysis provides a clear framework for understanding customer value.
  • By quantifying Recency, Frequency, and Monetary Value, we can effectively rank customers and identify those who are most valuable to the business.

High-Value Customer Engagement:

• Main insight 1: High-value customers ("DELIGHT" segment) require personalized and exclusive experiences.
  • These customers are highly engaged and contribute significantly to revenue. Maintaining their loyalty is crucial.
• Main insight 2: Personalized offers and VIP programs can enhance their loyalty.
  • Providing exclusive offers, early access to new products, or dedicated account management can further strengthen their relationship with the company.
• Main insight 3: Consistent communication and recognition are key.
  • Regular engagement with these customers, acknowledging their value, can foster a long-term relationship.

Marketing Optimization:

• Main insight 1: Different customer segments respond to different marketing strategies.
  • A one-size-fits-all approach is not effective. Marketing efforts should be tailored to the specific needs and preferences of each segment.
• Main insight 2: "UPSELL" customers are ideal targets for increasing purchase value.
  • Offering product bundles, cross-selling complementary items, or implementing loyalty programs can encourage these customers to spend more.
• Main insight 3: "PAMPER" customers may respond to offers on high-value items.
  • Personalized recommendations or promotions on luxury products can incentivize these customers to make more purchases.
• Main insight 4: "RETAIN", "RE-ENGAGE", "NURTURE", and "REWARD" segments require their own marketing strategies.
  • These segments, identified in the notebook, also need tailored approaches, such as loyalty programs, win-back campaigns, and nurturing efforts, respectively.

Outlier Customer Behavior:

• Main insight 1: Outliers represent distinct customer behaviors that require specific attention.
  • Analyzing outliers separately from the main customer base provides valuable insights.
• Main insight 2: Monetary outliers ("PAMPER") are high spenders but infrequent buyers.
  • Strategies to increase their purchase frequency can significantly boost revenue.
• Main insight 3: Frequency outliers ("UPSELL") are frequent buyers with low spending.
  • Upselling and cross-selling strategies can increase their average order value.
• Main insight 4: Monetary & Frequency outliers ("DELIGHT") are the most valuable customers.
  • Maintaining their loyalty and maximizing their lifetime value is paramount.

Recommendations:

Based on the insights and findings above, the following recommendations are made for the marketing team:

• The "DELIGHT" segment represents the most valuable customers. Develop a VIP program with exclusive benefits, personalized services, and proactive communication to maintain their loyalty and maximize their lifetime value.
• The "UPSELL" segment consists of frequent buyers with low spending. Implement targeted upselling and cross-selling campaigns, bundle deals, and loyalty programs to increase their average order value.
• The "PAMPER" segment includes high spenders with infrequent purchases. Personalize offers and recommendations for high-value items, and explore strategies to increase their purchase frequency, such as exclusive events or early access to new products.
• The four additional customer segments (RETAIN, RE-ENGAGE, NURTURE, and REWARD) require tailored strategies. Develop specific marketing plans for each segment, focusing on retention, re-engagement, nurturing, and rewarding behaviors, respectively.
• Regularly monitor and analyze the performance of each customer segment. Track key metrics such as customer lifetime value, retention rate, and purchase frequency to evaluate the effectiveness of targeted marketing strategies and make data-driven adjustments.