A customer profiling method for personalized service in retail pharmacies

Customer analysis, which describes users through data-driven methods, has become a fundamental technology widely used in retail scenarios. In order to enhance personalized services in the pharmacy industry and increase sales through the use of big data technology. By analyzing the consumption data of pharmacy members, we aim to increase member stickiness and attract consumers. This article proposes a customer behavior modeling method based on the online consumption history and purchase records of pharmacy members. In this study, we first analyzed the purchase history and consumption behavior data of pharmacy members. Then, we combine the drug purchase duration and the drug purchase cycle with the RFM model that reflects the consumption value to extract the latest value label of the user, which is more effective than the traditional RFM model. Then, a pharmacy membership service purchase prediction model based on LSTM was established to optimize the retail strategy for each member's potential future purchases. Based on experiments on real pharmacy datasets, the model can accurately identify members' preferences and needs, enabling pharmacies to provide more personalized service recommendations.