電信業帳單代收業務之欠費管理機制研究-以小額代收為例

Abstract

隨著各項的創新服務問世，在四年前電信業者也開始導入小額消費的電信帳單代收服務，本研究旨在透過電信業者擁有之用戶資料，研究哪些顯著變數對用戶小額帳單代收欠費會有影響，研究方法以台灣某電信業者有使用google play帳單代收用戶為研究樣本，資料處理採用羅吉斯迴歸分析方法。經主要研究發現，得到結論如下： 一、本研究找出影響用戶欠費的顯著變數共8個。 二、透過本次研究的實證，電信業者未來可以參考信用風險模型的相關建置與運作模式，建立預測用戶欠費機率的信用風險模型。 三、電信業者亦應持續標準化建檔其所蒐集到的用戶行為資訊，以增加不同面相之顯著變數。 從本研究發現，電信產業擁有較銀行業更豐富的用戶資料及使用行為，只要能找出影響欠費的顯著因子越多，對於欠費用戶的管理即會更精準，此項發現值得讓未來電信產業制定風控政策時之參考。

關鍵詞: 帳單代收、信用風險、欠費機率、羅吉斯迴歸分析。

Under the trend of innovation, telecom operators have introduced Direct Carrier Billing (DCB) of Micropayments four years ago. With the advantage of user information, the purpose of this study was to examine what kinds of significant variables that affecting debt of DCB. Subjects are the Google Play DCB users of one of Taiwan’s telecom operator and Logistic Regression Model is adopted. The main results are summarized as follows: 1.Eight significant variables are identified that affecting debt of DCB. 2.It is revealed that telecom operator can take the advantage of Credit Monitor Model to proactively predict the risk of DCB debt. 3.It is suggested that information of user behavior should be managed in standard operating procedures in order to explore different variables. To conclude, with abundant resources of user information and their usage, the more variables are found by telecommunication companies the more accurate management of bad debt user could be. This study can be taken as reference for telecommunication industry when building risk management policy.

Keywords: Direct Carrier Billing, Credit Risk, Debt Rate, Logistic Regression Model.