تقسیم بندی مشتریان مخابرات بر اساس ارزش مشتری با استفاده از مدل درخت تصمیم گیری
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی|
|2603||2012||10 صفحه PDF||22 صفحه WORD|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 39, Issue 4, March 2012, Pages 3964–3973
پژوهش های مربوط
محاسبه ی ارزش مستقیم
محاسبه ی ارزش مشتری براساس درخت تصمیم گیری
چارچوب عمومی محاسبه ی ارزش مشتری
ارزش جاری مشتری
ارزش بلندمدت مشتری
چارچوب عمومی برای محاسبه ی ارزش مشتری
محاسبه ی وفاداری
محاسبه ی ارزش غیرمستقیم
یافته های تجربی
نتیجه ی مدل CART.
روابط سلسله مراتبی و وزن (اعتبار).
The more the telecom services marketing paradigm evolves, the more important it becomes to retain high value customers. Traditional customer segmentation methods based on experience or ARPU (Average Revenue per User) consider neither customers’ future revenue nor the cost of servicing customers of different types. Therefore, it is very difficult to effectively identify high-value customers. In this paper, we propose a novel customer segmentation method based on customer lifecycle, which includes five decision models, i.e. current value, historic value, prediction of long-term value, credit and loyalty. Due to the difficulty of quantitative computation of long-term value, credit and loyalty, a decision tree method is used to extract important parameters related to long-term value, credit and loyalty. Then a judgments matrix formulated on the basis of characteristics of data and the experience of business experts is presented. Finally a simple and practical customer value evaluation system is built. This model is applied to telecom operators in a province in China and good accuracy is achieved.
The telecom industry in China was restructured in 2008 when 3G licenses were finally granted to three mobile operators. Since then, competition has been intensified further. As a result, telecom operators are paying much more attention to high-value customers. The 80/20 rule points out that 80% of the profits come from the top 20% of profitable customers and 80% of the costs are incurred on the top 20% of unprofitable customers (Duboff, 1992 and Gloy et al., 1997). However, finding the top 20% customers is the crucial issue for the operators. It is believed that companies who can capture the top 20% customers will win the battle for the market. Traditionally, experience-based or ARPU (Average Revenue per User) method is widely-used to find the top 20% customers in China’s telecom industry. In general, customers whose ARPU is ranked in top 20% are customers whose usage value is in the top 20% bracket. However, such a method considers only the current and the historic profit, but not future revenue and customer lifecycle. So this method cannot effectively discover the real high-value customers. For instance, customer A and customer B have different ARPUs (A is 200, B is 150), and their indirect values (e.g., loyalty, credit, etc.) may also be significantly different (A is 0, B is 50), however they may have the same contribution to the company’s profit, i.e., 200 (Fig. 1). On the other hand, a pair of customer A and B may have the same ARPUs, but their costs to the company may be significantly different. Nevertheless, it can be noted that they may have different contribution to the company (A is 200, B is 80) (Fig. 2).
نتیجه گیری انگلیسی
With the development of customer relationship management, customer value becomes more and more important for business decisions. Ways to evaluate a customer’s value are demanded by many enterprises. As far as we know, there is no practical customer value evaluation system. Therefore, it is needed to develop a customer value evaluation system. Here we develop a simple and practical customer value evaluation system for the telecom industry. The model can be used to predict customer lifecycle when only demographic information is available in the company’s data. The customer’s contribution is evaluated and quantified to support decision-making of the enterprise. To compute loyalty and credit, we use the AHP method. Different from other models, the score of weights from AHP is given not only by experts’ experience but also the characteristics data. It overcomes the subjectiveness of the experts’ scoring to some degree. Due to conditional constraints, there are still some limitations in our research. We do not pay much attention to the advantage of our model compared with others. When computing the long-term value, we have used a relatively simple way to compute the monthly average long-term value, which needs some further analysis and discussion.