家電用品顧客價 值分析

Abstract

近年來顧客關係管理(Customer Relationship Management)的議題受到熱烈的討論與研究。企業投入大量資源，以期望增加對顧客的瞭解並與顧客建立良好關係，希望藉此能夠提高顧客滿意度(Customer Satisfaction)及顧客忠誠度(Customer Loyalty)，進而增加忠實顧客。而其中，顧客價值分析(Customer Value Analysis)是顧客關係管理中重要的基礎。對企業而言，從顧客所獲得的利益越高，代表顧客的價值越高。依據80/20 經營法則，企業百分之八十的獲利是由最有價值的百分之二十顧客所創造，而剩下百分之二十的獲利才由較低價值的百分之八十的顧客所貢獻。因此，若企業能區隔高低價值的顧客，將大部分資源放在照顧高價值的顧客上，將有助於吸引更多高價值的客戶成為忠實顧客，進而使企業獲得最大的經營效益。

本研究以家電用品顧客歷史交易資料進行顧客價值分析。首先採用80/20法則為基礎，區分出高價值顧客及低價值顧客。然後應用統計方法中的二元羅吉斯迴歸，建構出高價值顧客的預測模型。研究結果發現，影響顧客成為高價值顧客的顯著因素為性別、職業、年齡及購買次數；其中男性、非學生、50歲以上的顧客成為高價值顧客的機率相對較高，而購買次數也與成為高價值顧客的機率呈顯著正相關，整體模型的準確歸類率可達75.6%。可讓家電業者利用本模型準確找到高價值顧客，確實利用企業有限資源，進行顧客關係管理，以達到減少企業資源浪費，提升企業競爭力的目的。

In recent years, the issue of Customer Relationship Management has gaining a lot of discussion and research. Enterprises are investing many resources to increase understanding of customers and establish good relationship with customers, in a hope to raise customer satisfaction and customer royalty and eventually to obtain more royal customers. Thereof, Customer Value Analysis is fundamental to CRM. From the perspective of an enterprise, a customer has a higher value if the enterprise gains more profit from the customers. According to the 80/20 rule, an enterprise could gain 80% of earnings on 20% of customers with higher value and the rest of earnings on 80% of customers with lower value. Accordingly, enterprises should discriminate the high value customers and low value customers. Putting most of the resources on customers with higher value will help to attract more high-valued customers to become royal customers, so as to gain maximum profit for the enterprise.

In this thesis, we use customer history transaction data of electrical appliances to perform customer value analysis. First, based on the 80/20 rule and discriminated the high value customers and low value customers. Then, using Logistic Regression to build the model of prediction to forecast the customer which type of customer was belongs to. The conclusion of this study may shows: the notable factors that make customers into high value ones are gender, profession, age and purchasing frequencies. The ratio that male, non-students, and the above 50 years old become high value customers is comparably high. And purchasing frequencies are positively relative with the ratio of becoming high value customers. The accurate rate of entire model can live up to 75.6%. It can be more accurately that let the electrical appliances company use this research model to find the cluster of high value customer, helps the company truly to carry on the customer using the enterprise limited resources to the customer relationship management, and achieves the goal that they can reduced the wasting of resources in the enterprise, and promote enterprise's competitive ability.