Approach
The segmentation model groups customers using the RFM (Recency, Frequency, Monetary) framework:
- Recency (R): Time since the last purchase
- Frequency (F): Number of purchases
- Monetary (M): Total amount spent
Using the RFM model, customers are categorized into segments such as Champions, Loyal Customers, Can’t Lose, Promising, About to Sleep, Hibernating, and New Customers.
Methodology
The KMeans algorithm is used for clustering, where unique centroids are randomly selected to group customers. A K-Nearest Neighbors (KNN) classifier is then trained to refine the segmentation by grouping closely related data points together
This approach helps the customer analytics team gain a clearer understanding of retail customer behaviors and supports targeted engagement strategies.
