Wine Customer Analysis

A comprehensive data analysis project focused on understanding customer behavior, spending patterns, and campaign effectiveness for a wine retail company.

📊 Project Overview

This repository contains a series of analytical notebooks that explore customer data to derive actionable business insights. The analysis progresses from basic exploratory data analysis through advanced segmentation and clustering techniques.

📁 Dataset

File: dataset/customer_analysis.csv

• Size: 2,240 customers × 29 attributes
• Content: Customer demographics, purchase behavior, spending patterns, and marketing campaign responses
• Documentation: See dataset/dictionary.md for detailed attribute descriptions

Key Data Categories

• Demographics: Age, education, marital status, income, household composition
• Products: Spending on wines, fruits, meat, fish, sweets, and gold products
• Promotion: Campaign acceptance rates and deal purchases
• Place: Purchase channels (web, catalog, store) and web visits

📓 Analysis Notebooks

The analysis is organized into sequential notebooks, each focusing on specific aspects:

0_customer_eda.ipynb - Exploratory Data Analysis

Initial exploration of the dataset covering:

• Data quality assessment and validation
• Demographic profiling
• Spending pattern analysis
• Purchase behavior insights
• Campaign performance overview
• Key findings and business recommendations

1_correlation_study.ipynb - Correlation Analysis

Statistical correlation analysis to identify:

• Relationships between customer metrics
• Strong positive and negative correlations
• Feature dependencies
• Insights for feature selection

2_rfm_analysis.ipynb - RFM Segmentation

Customer segmentation using RFM (Recency, Frequency, Monetary) methodology:

• RFM score calculation
• Customer tier classification
• Behavioral segment identification
• Targeted marketing recommendations

3_kmeans_clustering.ipynb - K-Means Clustering

Unsupervised machine learning for customer segmentation:

• Optimal cluster determination
• Customer group profiling
• Cluster characteristics analysis
• Segment-specific strategies

4_impact_analysis.ipynb - Impact Analysis

Feature importance and impact assessment:

• Key drivers of customer behavior
• Predictive feature analysis
• Business metric relationships
• Strategic recommendations

📚 Libraries Used

• pandas - Data loading and manipulation
• NumPy - Fundamental numerical operations (underneath pandas)
• matplotlib & seaborn - Data visualization
• scikit-learn - Machine learning (specific to clustering notebook)

📦 Prerequisites

To re-run the notebooks locally:

pip install pandas numpy matplotlib seaborn scikit-learn jupyter

📈 Key Insights

The analysis reveals:

• Wine products dominate customer spending
• Low campaign acceptance rates suggest need for better targeting
• Store purchases are the preferred channel
• Distinct customer segments with different behavioral patterns
• Opportunities for personalized marketing and product recommendations

For detailed findings and recommendations, refer to the individual notebooks.

📝 Project Structure

wine-customer-analysis/
├── dataset/
│   ├── customer_analysis.csv
│   ├── customer_data_segmented.csv
│   └── dictionary.md
├── 0_customer_eda.ipynb
├── 1_correlation_study.ipynb
├── 2_rfm_analysis.ipynb
├── 3_kmeans_clustering.ipynb
├── 4_impact_analysis.ipynb
└── README.md

🎯 Use Cases

This analysis can help businesses:

• Identify high-value customer segments
• Optimize marketing campaign targeting
• Improve customer retention strategies
• Develop personalized product recommendations
• Allocate resources more effectively across channels
• Create data-driven pricing and promotional strategies

📄 License

This project is available for educational and analytical purposes.

👤 Author

---

For questions or suggestions, please open an issue in this repository.