整合RFM和CAI工具分析顧客價值──超市大數據資料應用

Integrating RFM and CAI Tools for Analyzing Customer Value: The Application of Supermarket Big data

隨著大數據時代來臨，國外知名零售商店（如Target, Walmart）均已善用大數據資料，提升顧客關係行銷能力。然而，國內超市利用顧客資料庫進行顧客關係行銷能力較為不足，且過度著重於探討環境面因素對顧客購物準則之影響，忽略資料庫中的顧客消費數據與未來顧客購物間之相互依存關係。再者，大數據行銷分析工具之預測顧客行為能力及其衍生行銷策略是否奏效，端視於企業對顧客價值的了解與掌握，給予顧客適當標籤，有利提升後續的行銷精準度。本研究針對國內某連鎖超市之4,377位會員交易資料作為研究對象，檢驗RFM和顧客活躍性指標對未來顧客購買行為之預測能力，以及檢驗整合RFM和CAI方法建構FCAI指標進行顧客價值分群之效果。結果顯示，RFM和CAI均為有效之顧客行為預測工具；而整合此二工具建構之FCAI指標，將顧客價值區分為四種集群，其解釋能力較RFM和CAI指標更強，能更清楚地協助企業識別顧客價值。篇末提供行銷策略建議與未來研究方向。

In the big data era, well-known international retail stores (e.g., Target, Walmart) all utilize big data to enhance the capability of customer relationship marketing. However, domestic supermarkets utilize database to implement customer relationship marketing capabilities relatively insufficient, and they excessive focus on exploring the impact of environmental factors on the customers purchasing criterions, and not to analyze the database concerning the relationship between the current consuming data and the future consuming. In addition, the efficiency of the big data marketing analysis tools on predicting customer behavior and evaluating the derivative marketing strategy is up to the understanding of the customers. Customers shall be properly labeled to ensure the future marketing accuracy. The study analyzed the transaction data of 4,377 members of a domestic chain supermarket, examining the forecasting performance of the RFM and CAI index on future purchasing behavior of customers, and examining the efficiency of the FCAI index integrating RFM and CAI on the customer value clustering. The results indicate that RFM and CAI are effective forecasting tools on predicting future customer behavior, and the FCAI index integrating these two tools, dividing the customer value into four clusters, is a more powerful explanatory tool, assisting enterprise to identify customer value more clearly. Finally, we propose marketing implications and future research suggestions.