研究者鑒於零售商在退貨率的攀升以及其他廠商的競爭,迫使其必須制定更嚴格的退貨政策,故研究者在本篇論文中,試圖以電視購物產業為例,探討如何在「增加客戶滿意度的目標」以及「降低產品退貨率的迫切性」的兩難當中,求取一個平衡。 首先,作者檢討過去研究的一個預測模型 ──使用過去消費行為,來預測出較高退貨率的消費者;其次, 本文的目的在於建立篩選無獲益可能顧客的機制以及降低退貨率的管理方案。本研究試圖分析上述預測模型,並深入討論以下兩個面向: 第一,探討使用此模型時可能會面對的風險;第二,使用此模型評等顧客時,其他仍須涵蓋的相關變數。
Motivated by the fact that retailers’ products return rate is increasing while on the other hand they are squeezed by competition that harden them to pose stricter policy, then through this thesis, the author will try to explore possible strategies to balance retailers’ goal in increasing customers’ satisfaction and retailers’ necessity in decreasing products return rate by focusing on TV shopping industry. While most of previous researches were focusing more into improving return handling management, the strategies proposed here will highlight techniques to prevent return to happen at the first place, whether it was caused by fraud or general reasons, by using return policy and improving other internal and external activities. The author will first analyze a previously-researched prediction model which aimed to predict the high rate product returners based on their previous behavior, the analysis covers several topics such as identifying risks that might be faced when implementing the project, together with what other variables should be included when scoring customers. After that, appropriate ways to demarket unprofitable customers and what managerial actions could be implemented to decrease the overall return rate will be explored. The study shows that retailers are partly responsible for the high products return rate problem, and thus should alter parts that are still controllable to decrease it (through marketing strategy, product presentation, etc). We also found that lack of social presence could be both disadvantage and advantage for virtual retailers: it is one of the triggers of high products return rate, but it also enable retailers to implement many kinds of policies which are sometimes difficult to be implemented by brick-and-mortar retailers.