🚀 Customer Analytics & Machine Learning Project (Stripe Data)

Overview

This project focuses on customer data analysis and machine learning using Stripe payment data (300+ clients, 3000+ transactions). The goal was to extract insights, predict customer behaviour, and support business decisions such as location expansion and customer targeting.

​

📍 Geospatial Customer Analysis (Python)

Using Python, Pandas, and Geopy, customer location data was transformed into interactive maps to visualise geographic distribution.

​

This allowed the business to:

• Identify high-density customer areas

• Compare customer locations with existing sites

• Evaluate new location opportunities based on real data

​

This is an example of geospatial data analysis in Python applied to business strategy.

​

📊 Customer Behaviour & Data Analysis

Customer data was analysed using Python (Pandas, NumPy, Seaborn) to understand patterns in:

• Customer lifetime (months active)

• Total spend

• Payment frequency

• Churn, refunds, and active status

​

Statistical methods used:

• KDE plots and distribution analysis

• Cross-tab analysis (postcode vs time/day/class type)

• Chi-squared testing

​

This stage focused on customer behaviour analysis and identifying high-value clients.

​

🤖 Machine Learning for Customer Prediction

Machine learning models were built using Scikit-learn to predict customer outcomes:

• Logistic Regression

• Random Forest Classifier

​

These models were used to:

• Predict customer retention and churn

• Identify high-value customers

• Support data-driven decision making

​

This demonstrates practical machine learning applied to real-world customer data.

​

📈 Business Impact

• Improved understanding of customer lifetime value (CLV)

• Identified high-performing locations for expansion

• Reduced risk in business decision-making

• Enabled targeted marketing strategies

​

🧰 Tech Stack

Python, Pandas, NumPy, SQLite, Geopy, Matplotlib, Seaborn, Plotly, Scikit-learn

​