An Association Between Product Discount, Rating and Customer Service: An Empirical Study

Abstract

This study investigates the relationship between buyers' perceptions and empirical data regarding product ratings and discount percentages during e-commerce sales events. Through a mixed-methods approach combining survey analysis and real transactional data, the research reveals a significant discrepancy between consumer beliefs and actual trends. Survey results indicate that the buyers perceive a strong influence of product ratings on discount percentages. However, quantitative analysis of sales data contradicts this perception, demonstrating no statistically significant association between product ratings and discount levels. Instead, the study identifies a moderate positive correlation between customer service quality, measured by chat response time, and product ratings, suggesting that responsive support enhances customer satisfaction and, consequently, product evaluations. Additionally, the research explores three predictive models—linear regression, decision tree, and random forest—to forecast discount percentages based on historical sales patterns and consumer behavior metrics. These models provide actionable insights for e-commerce platforms to refine dynamic pricing strategies, optimize promotional campaigns, and improve overall customer experience. The findings highlight the importance of data-driven decision-making in e-commerce, as consumer intuition may not always align with actual market trends. By leveraging predictive analytics, retailers can better anticipate demand fluctuations, tailor discounts effectively, and foster greater trust through transparent pricing practices. This study contributes to the growing body of knowledge on consumer psychology and pricing optimization in digital marketplaces.