北京地区游戏充值客户分级评价数据

在游戏行业中，充值客户的行为数据（最近一次充值时间、充值频率、充值总金额）是衡量用户价值的核心指标。采集北京地区游戏充值客户的最近一次充值时间R(天数)、充值频次F（次数）、充值总金额M（元)，采用RFM模型对客户进行分级评价。用RFM分析方法把客户分为ABCD四级，对细分过后的不同客户采取相应营销策略，进行精准有效的促销推送。通过 RFM 划分客户群体后，游戏运营可实现：
1、精准营销：给核心A级客户推送高端活动，给D级流失客户推送大额福利券。
2、资源倾斜：优先保障高价值A级客户的服务，设置VIP专属客服。
3、风险预警：对最近一次充值时间R 值大的客户提前回访，防止流失。
RFM客户分级评价可为同行业企业管理不同等级的客户，实现精准个性化服务提供数据支持。数据处理：1、对从北京地区采集到的数据进行脱敏、降噪、清洗、聚集、分析。2、数据加工：基于RFM模型，结合客户最近一次充值时间R(天数)、充值频次F（次数）、充值总金额M（元)对客户进行打分，得到：R得分、F得分、M得分。 用percentrank函数对客户最近一次充值时间R(天数)、充值频次F（次数）、充值总金额M（元)进行评分，（a）、最近一次充值时间R(天数)间隔最短的客户排在最上面。按照从1-5评分，前20%的客户得5分，接下来的20%用户得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）、根据客户充值频次F（次数）从高到底依次对用户进行评分，前20%的客户在充值频次F（次数）的分数为5，以此类推。 (c)、根据客户充值总金额M（元)，前20%的客户在充值总金额M（元)的分数为5，以此类推。充值总金额M（元)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。3、通过对客户进行分级管理，满足不同等级客户的个性化需求， R、F、M 三个维度相互关联，共同构成对游戏充值客户的立体画像。三者结合可精准划分客户层级，为游戏行业分级运营、提高营销效率提供数据支撑。

In the gaming industry, behavioral data of paying customers (last recharge time, recharge frequency, total recharge amount) serves as a core metric for measuring user value. This dataset collects data on last recharge time R (in days), recharge frequency F (number of times), and total recharge amount M (in yuan) of gaming paying customers in the Beijing area, and applies the RFM model for hierarchical customer evaluation.

Using the RFM analysis method, customers are divided into four tiers: A, B, C, and D. Corresponding marketing strategies are adopted for each segmented customer group to deliver precise and effective promotional outreach. After segmenting customer groups via RFM, game operators can achieve the following objectives:
1. Precision Marketing: Push high-end activities to core Class A customers, and distribute large-value welfare coupons to Class D churned customers.
2. Resource Allocation: Prioritize service support for high-value Class A customers, and set up VIP exclusive customer service channels.
3. Risk Early Warning: Conduct early follow-up visits with customers with a large R value (last recharge time) to prevent customer churn.

RFM-based hierarchical customer evaluation can provide data support for enterprises in the same industry to manage customers across different tiers and deliver precise personalized services.

### Data Processing
1. Data Cleansing & Preprocessing: Desensitize, denoise, clean, aggregate and analyze the data collected from the Beijing area.
2. Scoring & Data Enrichment: Based on the RFM model, score customers using their last recharge time R (in days), recharge frequency F (number of times), and total recharge amount M (in yuan) to derive R Score, F Score, and M Score. The PERCENTRANK function is used to generate scores for the three metrics:
(a) Customers with the shortest interval of last recharge time R (in days) are ranked first. Scoring is conducted on a 1-5 scale: the top 20% of customers receive 5 points, the next 20% receive 4 points, the subsequent 20% receive 3 points, the following 20% receive 2 points, and the last 20% receive 1 point.
(b) Users are scored based on their recharge frequency F (number of times) from highest to lowest: the top 20% of customers get 5 points for recharge frequency F, and so on for the remaining tiers.
(c) Based on total recharge amount M (in yuan), the top 20% of customers receive 5 points for total recharge amount M, and so on. The 20% of customers with the lowest total recharge amount M receive 1 point.

The overall RFM score is calculated as: $RFM Score = 0.3 imes (R Score) + 0.3 imes (F Score) + 0.4 imes (M Score)$
Customers with a score ≥4 are classified as Class A, those with 3 ≤ score <4 as Class B, 2 ≤ score <3 as Class C, and those with score <2 as Class D.

3. Tiered Customer Management: Hierarchical management of customers meets the personalized needs of customers across different tiers. The three dimensions of R, F, and M are interrelated and jointly form a three-dimensional portrait of gaming paying customers. The combination of these three dimensions enables accurate segmentation of customer tiers, providing data support for tiered operations in the gaming industry and improving overall marketing efficiency.

搜集汇总

背景与挑战

背景概述

该数据集包含667条北京地区游戏充值客户记录，基于RFM模型（最近一次充值时间、充值频次、充值总金额）进行客户分级评价，分为A、B、C、D四级，旨在通过精准营销和资源优化提升游戏行业运营效率。数据每季度更新，采用Excel格式，适用于企业客户管理和个性化服务支持。