Tuesday, May 11, 2021
This paper investigates when personalized price discount is more efficient; i.e., to understand the heterogeneous impacts of personalized price discount given different discount conditions. Specifically, we focus on two essential factors pertaining to the discount conditions: discount depth and variation. Leveraging large-scale data sources to employ personalized price discount strategies has been prominent in the retail industry. It is thus inherently important for decision-makers to understand when personalized price discount can be more efficient. To answer our research question, we obtain unique access to a large data set from one leading on-demand platform in China. Employing a quasi-experiment setting, we utilize the difference-in-differences method coupled with matching to estimate the impact of personalized discount under different discount depth and variation conditions. We first uncover that personalized price discount not only increases firm revenue but also enhances consumer satisfaction, as compared to the mass discount with a same average discount depth for all consumers. We also find that a higher discount depth reduces the impact of personalized discount on firm revenue, while increasing the impact on consumer satisfaction. Therefore, if the decision-maker can only offer a low discount depth, adopting a personalized discount strategy would be especially valuable. Finally, we find that adding a bit discount variation can lead to a statistically significant improvement in firm revenue; however, when the discount variation further increases, the additional improvement diminishes. In addition, a higher discount variation increases consumer satisfaction. Our paper aims to help decision-makers better understand the impact of personalized discount under different discount conditions. Moreover, our empirical results provide important evidence for theoretical studies on revenue management and personalized discount.