Machine Learning-Based Customer Segmentation for Sustainable Marketing Strategies

Sustainable marketing addresses consumers whose environmental values, willingness to pay a premium and price sensitivity vary widely, and campaigns that treat them uniformly waste effort on the unreceptive and under-serve the committed. This paper applies machine-learning clustering to segment consumers on four attributes, environmental values, willingness to pay a premium, price sensitivity and brand engagement, and evaluates the segments as a basis for targeting sustainable offerings. In a simulation of 3,000 consumers drawn from four latent segments, k-means recovers the structure with an adjusted Rand index of 0.87; silhouette analysis favours two to four clusters, and four are retained on interpretability. The segments respond to a sustainable campaign at rates from 5% to 20%; targeting the two most responsive, which comprise 55% of consumers, yields a response rate of 18% against 12% for mass marketing, a lift of 47%. The paper interprets the segments as committed, engaged-premium, price-constrained and disengaged consumers, derives a differentiated strategy for each, and discusses the data, validation and ethical requirements for segmentation in sustainable marketing, including the risk that targeting the already committed neglects the larger task of moving the uncommitted.