Overview: This project, completed for AXANTEUS, a market research agency, focuses on segmenting consumers based on purchase behavior and promotional responsiveness. Using a dataset of 600 consumer profiles, I applied clustering and predictive modeling techniques to understand brand loyalty, identify value-conscious consumers, and optimize promotional targeting.
Objectives:
Segment consumers based on purchasing patterns and motivation.
Predict brand loyalty using machine learning models.
Identify value-conscious consumers to guide targeted marketing.
Tools & Techniques:
R Programming: Data preprocessing, K-means clustering, Random Forest modeling, and data visualization.
Statistical Imputation: Handled missing/zero values in variables like education, affluence index, and household size.
Feature Engineering: Normalized continuous variables, created custom indicators (e.g., value-consciousness), and calculated derived metrics like average price, transactions per brand run, etc.
Analytical Process:
K-Means Clustering: Used to identify behavioral and promotional segments using purchase volume, brand runs, and product preferences. Elbow method and silhouette scores guided optimal cluster selection.
Predictive Modeling: Built Random Forest and Logistic Regression models to classify value-conscious consumers and predict brand loyalty (brand runs).
Feature Importance: Identified key predictors like the affluence index, the number of transactions, and brand-specific volume shares influencing purchase decisions.
Outcomes:
Discovered three distinct consumer segments with varying brand loyalty and value sensitivity.
Achieved high sensitivity (96.7%) in identifying value-conscious consumers using logistic regression.
Provided actionable insights to help AXANTEUS clients design cost-effective promotional strategies and loyalty programs.
Deliverables:
Comprehensive Project Report
Well-Commented R Code for cleaning, clustering, and modeling
Understanding nuanced consumer behavior helps businesses improve ROI on marketing spend, design better campaigns, and boost retention. This project bridges demographic and behavioral data to support strategic decision-making.