Price is a core influence for purchase decisions, yet customers differ in how they respond to price variation. While certain customers exhibit strong sensitivity to price variations, others associate quality with higher product prices. Understanding, accurately measuring, and incorporating this heterogeneity of price sensitivity is essential for delivering personalized product recommendations and improving marketing efficiency. This study proposes two approaches for modeling consumer behavior in terms of price sensitivity. The first approach employs Gaussian Kernel Density Estimation (KDE) to analyze individual customer purchase patterns. The second follows a data-driven classification framework to categorize both products and customers, based on price levels, and uses this information to re-rank the recommendation products. Although both successfully re-rank the products to be recommended, the data-driven approach has superior performance in precision and recall, approximately 86%, against approximately 64% for the KDE solution. Preliminary findings suggest that incorporating these components into a base recommendation system returns more relevant and personalized results, aligning with the customer’s price preferences and thereby increasing conversion rates.