Dynamic Pricing Method in the E-Commerce Industry Using Machine Learning

One of the key areas of contemporary marketing is the formulation of a pricing strategy, which is one of the four pillars of the traditional marketing mix. One way to implement this strategy is through dynamic pricing. It is currently gaining popularity in many industries for two reasons. Firstly, it is possible, easy, and cheap to collect information about transactions and customers. Secondly, machine learning mechanisms, for which these data are essential, are becoming widely available. The aim of this article is to propose a dynamic pricing method for the e-commerce industry. To achieve this goal, machine learning methods such as the Naive Bayes classifier, support vector machines (linear and nonlinear), decision trees, and the k-nearest neighbor algorithm were used. The empirical results indicate that the linear support vector machine achieved the highest accuracy (86.92%), demonstrating the model’s effectiveness in classifying pricing decisions. This article aligns with two leading research trends in dynamic pricing: personalized dynamic pricing (the target model considers customer-related criteria) and the development of systems to assist managers in optimizing pricing strategies to increase revenues (using machine learning methods). This article presents a literature review on dynamic pricing and then discusses the machine learning methods applied. In the final part of this article, verification of the developed dynamic pricing method using real-world conditions is presented.