Oacquah31/Payswift

Data-driven analysis for PaySwift Ghana to understand customer churn, predict loan default risk, and improve digital feature adoption. Includes segmentation, risk modeling, and actionable insights powered by Python, SQL, and Power BI. # Payswift Data-driven analysis for PaySwift Ghana to understand customer churn, predict loan default risk, and improve digital feature adoption. Includes segmentation, risk modeling, and actionable insights powered by Python, SQL, and Power BI. # 💼 PaySwift Customer Behavior Analysis **A data-driven project for segmentation, feature adoption, and risk modeling to improve retention and reduce loan defaults at PaySwift Ghana.** --- ## 🧭 Project Overview PaySwift Ghana is a fintech company offering mobile payments, micro-loans, and personal finance tools. The company faced critical challenges: - 📉 Low adoption of digital features - 💸 High loan default rates - 🚪 Customer churn shortly after onboarding Using real user data, this project explored how feature usage, repayment behavior, and transaction frequency influence churn and loan performance — leading to actionable strategies for growth and risk reduction. --- ## 🎯 Objectives - Identify behavioral drivers of user **engagement** and **churn** - Predict **loan default risk** based on usage patterns - Segment customers into meaningful **personas** - Recommend improvements in **UX**, **product strategy**, and **retention** --- ## 📘 Data Dictionary | Feature | Description | Type | |---------------------------|----------------------------------------------------------|-------------| | Customer ID | Unique user identifier | Categorical | | Feature Usage | Use of tools like Budgeting, Loans, Savings | Categorical | | Loan Repayment Behavior | On-time, Late, or Defaulted repayment status | Categorical | | Churn Flag | Indicates whether a user exited post-onboarding | Binary | | Transaction Frequency | Number of monthly transactions | Numeric | | Feature Type | Financial tool type (e. …