基于RFM模型的包装盒客户分级评价数据

在公司主要产品包装盒的经营销售领域中，为了更好地理解客户采购行为，以提高客户满意度和企业收益。通过收集客户的消费记录，使用RFM客户价值模型，用户最近一次消费时间(（Recency）、消费频率（Frequency）和消费金额（Monetary）进行评分，识别不同价值的客户群体。为客户定制个性化的营销和服务方案、提高客户满意度和忠诚度、增加客户留存率和生命周期价值。RFM模型通过计算客户最近一次消费时间(R)、消费频率(F)和消费金额(M)这三个维度来评估用户价值。对于R维度,根据客户最近一次消费距离分析日期的天数（D）,划分为5个等级: 0≤D≤30为5分，30＜D≤60为4分，60＜D≤90 为3分，90＜D≤120为2分，D ＞120为1分；对于F维度,根据用户在最近一年内的消费次数（C）,划分为5个等级: C≥8为5分、6≤C≤7为4分、4≤C≤5 为3分、2≤C≤3 为2分、0≤C≤1为1分；对于M维度,根据用户在最近一年内的消费金额（G）,划分为5个等级，G≥30000为5分，25000≤G＜30000为4分，20000≤G＜25000为3分，15000≤G＜20000为2分，G＜15000为1分。RFM综合评分（X）=R+F+M，再根据RFM综合评分（X）对客户进行分类，0≤X＜2为一般客户，2≤X＜4为新客户，4≤X＜6 为潜力深耕客户，6≤X＜8为重要维系客户，X ≥8为高粘度客户，基于消费频次、消费金额等不同维度获得的聚类分组成果，对聚类分组数量和分组阀值、以及维度权重进行人为干预，使客户分类趋于合理。

In the field of sales and operations for the company's core product packaging boxes, this work aims to better understand customers' purchasing behaviors to improve customer satisfaction and corporate revenue. By collecting customers' consumption records and applying the RFM customer value model, we score customers across three standard dimensions: Recency (time since last purchase, R), Frequency (purchase frequency, F), and Monetary (total consumption amount, M), to identify customer segments with distinct value profiles. This allows us to develop personalized marketing and service plans for customers, thereby boosting customer satisfaction and loyalty, as well as increasing customer retention rate and lifetime value.

The RFM model assesses customer value by analyzing three core dimensions: Recency (R), Frequency (F), and Monetary (M). For the Recency (R) dimension, customers are divided into 5 tiers based on the number of days (D) between their last purchase and the analysis date: 5 points for 0≤D≤30, 4 points for 30<D≤60, 3 points for 60<D≤90, 2 points for 90<D≤120, and 1 point for D>120. For the Frequency (F) dimension, customers are divided into 5 tiers based on the total number of purchases (C) made by the user within the most recent year: 5 points for C≥8, 4 points for 6≤C≤7, 3 points for 4≤C≤5, 2 points for 2≤C≤3, and 1 point for 0≤C≤1. For the Monetary (M) dimension, customers are divided into 5 tiers based on the total consumption amount (G) of the user within the most recent year: 5 points for G≥30000, 4 points for 25000≤G<30000, 3 points for 20000≤G<25000, 2 points for 15000≤G<20000, and 1 point for G<15000.

The comprehensive RFM score (X) is calculated as X = R + F + M. Customers are then classified based on their comprehensive RFM score (X): general customers for 0≤X<2, new customers for 2≤X<4, potential in-depth development customers for 4≤X<6, key retention customers for 6≤X<8, and high-stickiness customers for X≥8. Based on the clustering results derived from multiple dimensions such as purchase frequency and consumption amount, manual adjustments are made to the number of clustering groups, grouping thresholds, and dimension weights to optimize the rationality of customer classification.