武汉市舟山旅游客户分级评价数据
收藏资源简介：
通过收集和分析武汉市游客在企业云商平台上对舟山旅游产品（景区门票、船票、大巴票、机票等）消费相关数据，使用RFM客户价值模型，了解游客对舟山旅游的购买力水平和消费偏好，对游客进行等级评级，实现精准化运营。通过对客户价值管理，满足不同价值客户的个性化需求。对于A等级客户可每月1至2次推送优惠信息与之沟通，对于B等级客户可每季度1至2次发送旅游资讯与之沟通，对于C等级客户可每半年1至2次推送活动信息与之沟通。另外可以为舟山本地旅游相关企业（景区、酒店、旅行社、交通公司等）提供不同等级客户个性化服务的数据支持。1.数据采集：采集武汉市游客在企业云商平台上对舟山旅游产品（景区门票、船票、大巴票、机票等产品）的相关交易数据。其中，采集数据中"下单时间"为距离统计时间（2025/8/31）最近的一次订单时间，"订单金额"指的是距离统计时间最近的这次订单金额，"历史购买总次数""历史购买总金额"指的是历史服务时间段（2024/1/1-2025/8/31）内统计得出的购买次数和购买金额。 2.数据处理：对采集到的本次订单金额（元）、历史购买总金额（元）等数据进行分类、合并、累加，便于分析使用，其中客户编号已进行脱敏转换处理。 3.算法加工：R评分：根据用户下单时间距离统计时间的天数（D）划分为5个等级：0≤D≤10为5分，10＜D≤20为4分，20＜D≤30为3分，30＜D≤50为2分，50<D为1分；F评分：消费频率评分根据历史购买总次数（S），划分为5个等级：0<S≤2为1分，2＜S≤5为2分，5＜S≤10为3分，10＜S≤15为4分，15<S为5分；M评分：根据历史购买总金额（Z，单位：元），划分为5个等级：0<Z≤5000为1分，5000＜Z≤10000为2分，10000＜Z≤15000为3分，15000＜Z≤30000为4分，30000<Z为5分；RFM综合评分（X）=0.3×R+0.4×F+0.6×M；客户等级分为ABC三级，0≤X≤3为C级，3＜X≤6为B级，6＜X为A级。
By collecting and analyzing consumption data of Zhoushan tourism products (including scenic spot tickets, ferry tickets, bus tickets, air tickets, etc.) from Wuhan tourists on the enterprise cloud commerce platform, and adopting the RFM customer value model, this dataset aims to understand the purchasing power and consumption preferences of tourists for Zhoushan tourism, grade tourists, and realize precise operation. Through customer value management, personalized needs of customers at different value levels can be met. For Grade A customers, push preferential information to communicate with them 1 to 2 times per month; for Grade B customers, send tourism information to communicate with them 1 to 2 times per quarter; for Grade C customers, push activity information to communicate with them 1 to 2 times every half year. In addition, it can provide data support for personalized services of different customer tiers for local tourism-related enterprises in Zhoushan (including scenic spots, hotels, travel agencies, transportation companies, etc.). 1. Data Collection: Collect relevant transaction data of Zhoushan tourism products (such as scenic spot tickets, ferry tickets, bus tickets, air tickets, etc.) from Wuhan tourists on the enterprise cloud commerce platform. In the collected data, "order time" refers to the time of the most recent order relative to the statistical time (2025/8/31), "order amount" refers to the amount of the most recent order relative to the statistical time, and "total historical purchase times" and "total historical purchase amount" refer to the total purchase times and total purchase amount counted within the historical service period (2024/1/1 – 2025/8/31). 2. Data Preprocessing: Classify, merge and accumulate the collected data such as the current order amount (unit: yuan) and total historical purchase amount (unit: yuan) to facilitate subsequent analysis. Customer IDs have been desensitized and converted. 3. Algorithm Processing: - R Score: Divide the number of days (D) between the user's order time and the statistical time into 5 tiers: 0≤D≤10 scores 5 points, 10<D≤20 scores 4 points, 20<D≤30 scores 3 points, 30<D≤50 scores 2 points, and D>50 scores 1 point; - F Score: The consumption frequency score is divided into 5 tiers based on the total historical purchase times (S): 0<S≤2 scores 1 point, 2<S≤5 scores 2 points, 5<S≤10 scores 3 points, 10<S≤15 scores 4 points, and S>15 scores 5 points; - M Score: Divide the total historical purchase amount (Z, unit: yuan) into 5 tiers: 0<Z≤5000 scores 1 point, 5000<Z≤10000 scores 2 points, 10000<Z≤15000 scores 3 points, 15000<Z≤30000 scores 4 points, and Z>30000 scores 5 points; - RFM Comprehensive Score (X) = 0.3×R + 0.4×F + 0.6×M; - Customer grades are divided into three levels: Grade C for 0≤X≤3, Grade B for 3<X≤6, and Grade A for X>6.
