All projects
Data Analysis2026

Customer Segmentation & Value Analysis

RFM + K-Means segmentation isolating the high-value cohorts that drive revenue.

PythonPandasK-MeansPower BISeaborn
0

RFM dimensions

0

Customer segments

0%+

Top-cohort revenue

0+

Power BI dashboards

Overview

An unsupervised analytics pipeline that segments customers from transactional data using K-Means clustering on engineered RFM features. High-value cohorts are surfaced and communicated through Power BI dashboards to support targeted marketing and retention.

Features

  • Unsupervised K-Means clustering on transactional data
  • RFM feature engineering (Recency, Frequency, Monetary)
  • Improved cluster separability & business interpretability
  • High-value cohort isolation
  • Power BI dashboards for marketing & retention

Challenges

  • Choosing a cluster count that separated customers cleanly.
  • Making abstract clusters business-interpretable via RFM.
  • Turning segments into actionable marketing strategy.

Results

RFM-based K-Means produced clearly separated segments and isolated the high-value cohorts driving the majority of revenue — enabling targeted retention and marketing through the Power BI dashboards.

Screenshots

RFM feature distribution
RFM feature distribution
Cluster scatter projection
Cluster scatter projection
Power BI segment dashboard
Power BI segment dashboard
How it works

Architecture

The end-to-end flow from raw data to a shipped result.

01

Ingest

Load transactional data

02

Engineer

Build RFM features

03

Cluster

K-Means segmentation

04

Visualise

Power BI dashboards

Keep exploring

Other projects