南京地区智能窗帘消费客户分级评价数据

采集销售记录表中南京地区的数据，通过客户在2021年9月1日距离2025年7月1日间隔的最近一次消费时间天数R、客户在2021年9月1日至2025年7月1日之间消费频次F和客户在2021年9月1日至2025年7月1日之间消费金额M（单位：元）， 采用 RFM 模型对客户进行价值评级，实现精准化运营，通过对南京地区客户价值管理，满足不同价值客户的个性化需求。对A级客户，每个月进行一次回访维护，对B级客户，每个季度进行一次回访维护，对C级客户每半年进行一次回访维护，对D级客户每年进行一次回访维护。另外可以为本地区客户群体高度重叠企业提供不同价值类型的客户个性化服务的数据支持。对从销售记录表中采集到的数据进行脱敏、降噪、清洗、聚集、分析。2、数据加工：运用RFM模型结合客户在2021年9月1日距离2025年7月1日间隔的最近一次消费时间天数R、客户在2021年9月1日至2025年7月1日之间消费频次F和客户在2021年9月1日至2025年7月1日之间消费金额M（单位：元）的得分排名对客户进行一个综合排名，最终得出一个RFM总评分。a.提取出最近一次消费时间距离当前分析时间的天数R、客户在2021年9月1日至2025年7月1日之间消费频次F和客户在2021年9月1日至2025年7月1日之间消费金额M（单位：元）进行分类，最近一次消费时间间隔最短的客户排在最上面。按照从1-5评分，前20%的客户获得5分，接下来的20%用户获得4分,再下来20%的客户为3分，再下来20% 的客户为2分，最后20% 的客户为1分。 b.根据客户在2021年9月1日至2025年7月1日消费频次F从高到底依次对用户进行分类，前20%的客户在用户活动频率的分数为5，以此类推。 C, 根据客户在2021年9月1日至2025年7月1日消费金额(M)，前20%的客户在消费金额的分数为5，以此类推。消费金额最少的20%客户则分数为1。 RFM得分=0.3*（R得分）+0.3*（F得分）+0.4*（M得分） 评分大于等于4分的为A级客户，大于等于3小于4的为B级客户，大于等于2小于3的为C 级客户，低于2的为D级客户。

This dataset extracts sales data for the Nanjing region from the official sales record sheet. Three core metrics are calculated for each customer: Recency (R): the number of days between the customer's most recent purchase and the analysis cutoff date of July 1, 2025, within the reference period spanning September 1, 2021 to July 1, 2025; Frequency (F): the total number of purchases made by the customer during this reference period; Monetary (M): the total consumption amount of the customer during the same period, with the unit being Chinese Yuan (CNY). The RFM model is employed to segment customers by their value, enabling precise operational management. By managing customer value in the Nanjing region, personalized needs of customers across different value tiers can be met. For Class A customers, monthly return visits and maintenance are conducted; for Class B customers, quarterly return visits; for Class C customers, semi-annual return visits; and for Class D customers, annual return visits. Additionally, this dataset can provide data support for enterprises with highly overlapping local customer groups to deliver personalized services tailored to different customer value types.

The collected sales data undergo preprocessing steps including data desensitization, denoising, cleaning, aggregation and analysis.
2. Data Processing: A comprehensive customer ranking is generated by combining the score rankings of the three RFM metrics, and a final total RFM score is calculated.
a. Recency (R) Scoring: Customers are sorted in ascending order of the number of days from their last purchase to July 1, 2025. They are divided into 5 equal quintiles, with the top 20% (i.e., customers with the shortest recency) receiving a score of 5, the next 20% receiving 4 points, the subsequent 20% receiving 3 points, the following 20% receiving 2 points, and the last 20% receiving 1 point.
b. Frequency (F) Scoring: Customers are sorted in descending order of their total purchase count between September 1, 2021 and July 1, 2025. They are divided into 5 equal quintiles, with the top 20% receiving a score of 5, and the remaining groups assigned scores in descending order accordingly.
c. Monetary (M) Scoring: Customers are sorted in descending order of their total consumption amount between September 1, 2021 and July 1, 2025. They are divided into 5 equal quintiles, with the top 20% receiving a score of 5, and the bottom 20% (those with the lowest consumption amount) receiving 1 point, with the remaining groups assigned scores in descending order accordingly.

The total RFM score is calculated using the formula: RFM Score = 0.3 * (R Score) + 0.3 * (F Score) + 0.4 * (M Score). Customer value tiers are defined as follows: Class A customers with a total RFM score ≥ 4; Class B customers with 3 ≤ total score < 4; Class C customers with 2 ≤ total score < 3; and Class D customers with a total score < 2.

搜集汇总

背景与挑战

背景概述

该数据集包含南京地区566条智能窗帘消费客户记录，采用RFM模型（基于最近消费时间、消费频次和消费金额）对客户进行分级评价，分为A、B、C、D四个等级，用于支持精准化运营和个性化客户服务管理。