The feature-value paradox: Unsupervised discovery of strategic archetypes in the smartphone market using machine learning

This study employs unsupervised machine learning, a core branch of Artificial Intelligence (AI), and feature importance analysis to identify strategic archetypes in the smartphone market based solely on technical specifications. Moving be- yond traditional price prediction models, we analyze a comprehensive dataset to discover latent product strategies. Using K-Means clustering, we identify five distinct strategic archetypes, which we then analyze against price categories to re- veal both aligned and paradoxical positioning strategies. Our findings demonstrate that approximately 23% of devices exhibit a feature-value paradox, where premium specifications are not rewarded with premium pricing. Through permutation importance analysis, we quantify the feature importance driving each archetype. This research contributes to marketing science and engineering practice by offering a novel, AI-driven methodology for reverse-engineering product strategies, with direct implications for product portfolio optimization and competitive positioning in technology markets.