rachel93augesty-del/Customer-Experience-Analytics-for-Fintech-Apps-challenge--Week-2

A Real-World Data Engineering Challenge: Scraping, Analyzing, and Visualizing Google Play Store Reviews of three banks mobile banking app in Ethiopia # Customer Experience Analytics for Fintech Apps ## Banking Apps Review Analysis - Week 2 Challenge ### 📋 Project Overview This project analyzes customer satisfaction with mobile banking apps by collecting and processing user reviews from the Google Play Store for three Ethiopian banks: - **Commercial Bank of Ethiopia (CBE)** - **Bank of Abyssinia (BOA)** - **Dashen Bank** ### 🎯 Business Objective Omega Consultancy is supporting banks to improve their mobile apps to enhance customer retention and satisfaction. As a Data Analyst, my role is to: - Scrape user reviews from the Google Play Store - Analyze sentiment and extract themes - Identify satisfaction drivers and pain points - Store cleaned review data - Deliver actionable recommendations ### 📊 Task 1: Data Collection and Preprocessing - COMPLETED ✅ #### Data Collection - **Tool**: `google-play-scraper` Python library - **Target Apps**: - CBE: `com.combanketh.mobilebanking` - BOA: `com.boa.boaMobileBanking` - Dashen: `com.dashen.dashensuperapp` - **Sample Size**: 400+ reviews per bank (1,200+ total) - **Data Points Collected**: - Review text - Star rating (1-5) - Review date - Bank name - User name - Thumbs up count - App version #### Data Preprocessing Pipeline The preprocessing follows a structured 6-step approach: 1. **Missing Data Handling** - Removed rows with missing critical data (review text, rating, bank name) - Filled missing user names with 'Anonymous' - Filled missing thumbs up with 0 - Handled missing reply content and app versions 2. **Date Normalization** - Converted all dates to YYYY-MM-DD format - Extracted year and month for temporal analysis - Date range: 2024-09-26 to 2025-11-25 3. **Text Cleaning** - Removed extra whitespace and normalized text - Preserved Ethiopian languages (Amharic, etc.) - Focused on English text preparation for NLP - Calculated review length and word count metrics 4. **Rating Validation** - Ensured all ratings are within 1-5 scale - Converted to numeric format - …