以個別顧客活躍率之隨機選擇模型建立一對一行銷分析

Abstract

本篇研究欲就顧客關係管理的觀點出發，針對虛擬通路之顧客進行區隔，以利進行未來之一對一行銷。以建立發展完善的隨機模型方式，定義顧客之活躍率，並將此指標作為顧客區隔的標準，進而將顧客區分為三個族群：高活躍率、中活躍率與低活躍率之顧客，再就不同區隔內的顧客族群進行深入的分析探討，旨在研究具有高度活躍力的顧客其影響因素為何，進而延伸至公司之策略因應，以提升顧客忠誠與顧客活躍率，使得組織能夠達到績效極大化之表現，並將影響因素作為組織在打入目標族群市場的參考依據與策略擬定之建議。

Base on the view point of Customer Relationship Management, discriminant customers in the virtual buying channel has benefits for One-to-One Marketing. Doing discrimination from the well development stochastic choice model, SMC model, also defined the probability of customer active as the basis to do customer discriminant. Then cluster customers active into three groups: high, medium, and low. The customers in different clusters might have different behavior, and the purpose is to figure out what are the key factors to affect customers active, and the high active customers will affect company’s performance directly is the key group we concerned. In order to reach better performance, it needs to increase customers’ loyalty and their actives. Therefore, the information will have a great benefit for future One-to-One marketing and they can predict customers’ behavior more easily.