Considering the level of competition prevailing in Business-to-Consumer (B2C) E-Commerce domain and the huge investments required to attract new customers, firms are now giving more focus to reduce their customer churn rate. Churn rate is the ratio of customers who part away with the firm in a specific time period. One of the best mechanism to retain current customers is to identify any potential churn and respond fast to prevent it. Detecting early signs of a potential churn, recognizing what the customer is looking for by the movement and automating personalized win back campaigns are essential to sustain business in this era of competition. E-Commerce firms normally possess large volume of data pertaining to their existing customers like transaction history, search history, periodicity of purchases, etc. Data mining techniques can be applied to analyse customer behaviour and to predict the potential customer attrition so that special marketing strategies can be adopted to retain them. This paper proposes an integrated model that can predict customer churn and also recommend personalized win back actions.
翻译:考虑到商业对消费者(B2C)电子商务领域的竞争水平以及吸引新客户所需的巨额投资,各公司现在更加关注降低客户的消费率。Curn比率是特定时期内与公司分离的客户比率。保留现有客户的最佳机制之一是查明任何潜在的贸易量,并迅速作出反应加以防止。发现潜在贸易量的早期迹象,承认客户在运动中寻找什么,实现个人化赢回运动的自动化对于在竞争时代维持商业至关重要。电子商务公司通常拥有大量与现有客户有关的数据,如交易历史、历史搜索、采购周期等。数据挖掘技术可用于分析客户行为,预测潜在的客户消耗量,以便采用特殊的营销战略予以保留。本文提出了一个综合模型,可以预测客户的消费量,并建议个人化赢回行动。