miracleezekiel/ecommerce-sales-intelligence

This project is a data analytics portfolio case study built around a synthetic e-commerce dataset of 5,000 transactions and 4,844 customers spanning two years. It is not a piece of software you install and run, it is a documented walk through of how one analyst investigated a business question: why does the data show strong revenue but very few customers coming back to buy again. The work is organized into five completed phases. The first phase cleans and validates the raw data in Google Sheets, checking for missing values, duplicates, and calculation errors. The second phase builds key performance indicators, pivot tables, and charts to explore revenue by region, category, and time period. The third phase moves the data into MySQL and writes SQL queries to dig into segmentation and churn, finding that over 96 percent of customers placed exactly one order and never returned, while customers who ordered two or three times spent multiples more over their lifetime. The fourth phase uses Python for RFM customer segmentation, a method that scores customers by how recently, how often, and how much they buy, sorting them into groups like Champion, Loyal, At Risk, and Lost. The fifth phase compiles everything into a written case study with CRM recommendations for each customer segment. The repository is organized into numbered folders holding the raw and cleaned datasets, the Google Sheets workbook, SQL scripts, Python notebooks, screenshots, and the final case study report in both Markdown and PDF. The README notes the dataset is synthetic, so the low repeat purchase rate should be read as a demonstration finding rather than a real business conclusion, and the author added extra validation notes after community feedback pointed this out. Overall this is a learning and portfolio artifact for someone building skills in business intelligence, CRM analytics, and customer segmentation using Google Sheets, MySQL, and Python together.