Customer Segmentation in Python: RFM Scores and k-Means Clustering

WagWell Pet Supplies, a Bengaluru online pet store, has ₹6 lakh to spend on bringing customers back this year. The founder wants a 20%-off coupon for everyone. Nisha Rao, the growth lead, wants to know who the customers are first: who buys often, who has gone quiet, and who was never going to come back.Across five lessons you build that answer in Python on your own copy of two years of raw order lines. You clean the export (duplicate lines, cancelled and returned-to-origin orders, lines that are not orders, quantities), build the RFM table with a fixed analysis date, score recency, frequency and monetary by quintile, and name six segments with first-match-wins rules. Then you let the data speak: log-transform and standardise the features, choose k with an elbow chart and the silhouette score, fit k-means, and describe each cluster in days, orders and rupees.In every lesson you build the step yourself first, then give the same job to your AI (ChatGPT, Claude, Copilot or Gemini) and learn exactly where it goes wrong: counting lines as orders, the tutorial's "latest date plus one day", recency scored backwards, rules that overwrite each other, k-means on raw rupees, a profile in z-units. In the capstone your AI writes the whole segmentation from one brief, and you prove every number before the budget follows it.Bartman runs your script on its own servers against your data and checks the files it writes, customer by customer, so every check is about your real output, not a screenshot.

Tools: Python, pandas, scikit-learn, k-Means Clustering, RFM Analysis, Customer Segmentation, matplotlib, AI Assistants for Python