data analytics bi
CustomerLens BI
A hybrid BI + ML customer segmentation project combining SQL-based RFM analysis, Power BI dashboards, and an interactive Streamlit KMeans segmentation app.
PythonStreamlitSQLPower BIPandasScikit-learnKMeansPlotly
Project Overview
CustomerLens BI is a customer segmentation workspace that combines traditional business intelligence with applied machine learning. The project starts with SQL-based RFM analysis and Power BI reporting, then extends the workflow into an interactive Streamlit app where users can upload their own CSV files and generate customer segments.
Project Goal
- Analyze customer behavior with SQL-based RFM features.
- Build Power BI dashboards for executive overview, segment deep dive, and country-level marketing actions.
- Create an interactive Streamlit app that supports customer, transaction, and profile CSV files.
- Apply upload-specific KMeans clustering instead of forcing every dataset into a fixed pretrained model.
- Let users rename segments, inspect segment insights, and export segmented customer data.
Model Architecture
1.SQL pipeline prepares cleaned customer-level features and RFM-style business segments.
2.Power BI dashboards present business-facing customer segmentation insights.
3.Streamlit app handles CSV upload, dataset type detection, column mapping, feature engineering, clustering, visualization, and export.
4.KMeans clustering is fitted fresh on each uploaded dataset to respect dataset-specific customer distributions.
5.Silhouette score, segment size warnings, PCA visualization, and business-friendly labels help users interpret the results.