Research On The Decision Making Of Electronic Coupon Marketing For E-commerce Enterprises

This research starts from the formulation of preferential policies and the precise delivery of e-coupons in e-commerce enterprises,which researches the e-coupons delivery decisions from the two levels of e-coupons and users.From the perspective of e-coupons,this paper explores the influencing factors of the redemption rate of e-coupons through the extraction and analysis of an e-commerce enterprise's e-coupon promotional data in the past two years,and based on this we make a research on e-coupon pricing decision making model.Data analysis shows that no threshold coupons have a higher redemption rate than coupons with amount limits.There is a significant positive correlation between the e-coupon's face value and the redemption rate.The higher the e-coupon's face value,the higher the redemption rate is.When the e-coupon's face value is fixed,the higher the threshold value of the e-coupon,the lower the redemption rate is.The study of the e-coupon pricing decision model finds that there is no direct correlation between sales profit and e-coupon's redemption rate.In order to maximize sales profits,e-commerce enterprises need to take account of factors such as original price and production cost function when deciding the type and face value of e-coupons.From the perspective of users,based on user log data and transaction data,data mining method is used to build prediction models of e-coupon redemption behavior before/after issuance.Based on the prediction model of e-coupon redemption behavior before issuance,it is found that for users who have access behavior before the issuance of e-coupons,the history of discount usage behavior(the number of e-coupons used,etc.)and the characteristics of access behavior before coupon issuance(the number of access to product details within a month prior to the issuance of e-coupons,etc.)are related to coupons redemption behavior.Based on the prediction model of e-coupon redemption behavior after issuance,it is found that for users who have access behavior in the validity period of e-coupons,there is a correlation between the total amount of historical transactions,the behavior of historical preferential use and the characteristics of the access behavior in the validity period and the e-coupon redemption behavior.Through the evaluation of the models,it is found that the classification models have certain accuracy.Therefore,it is feasible to predict the e-coupon redemption behavior based on user log data and transaction data.