Purchase likelihood prediction for targeted organic food marketing campaigns in China

The demand for organic food products in China is growing in response to both increased spending power and food safety concerns. However, identifying likely buyers of organic products proves challenging due to their relatively small fraction in the overall population. Our study explores applications of machine learning algorithms for effective management of organic food marketing campaigns in China. Based on the data we collected through an online choice-experiment type questionnaire of Chinese consumers, a purchase likelihood estimation framework has been developed that utilizes customer profile traits such as age group, family status, education level, and geographic location. In addition, we apply clustering techniques to perform data-driven organic market segmentation and identify consumer profiles ready to pay more for high quality, certified organic products. The resulting market segments are compared to various types of organic consumers discussed in the literature. Our algorithms provide a useful framework for online retailers who are seeking lean strategies of market entry in China with their health food brands.