东阳共享电动车用户活跃度管理数据

通过对用户的骑行时间、骑行频率等行为数据进行分析，识别用户的骑行类型，并实时监测用户活跃度变化。当用户活跃度显著下降时，系统自动触发针对性的优惠推送，以提升用户留存率和复购率。该方案解决了传统用户运营中“一刀切”优惠策略的无效性问题，实现精准用户挽留，优化营销资源分配。1、数据来源：从东阳创享汽车服务有限公司的创享出行共享电动车运营平台系统中，采集用户数据（实名用户）和用户骑行订单数据。 2、数据处理：对采集的数据进行数据清洗，剔除异常订单数据（如骑行时长小于1分钟或大于60分钟的订单数据等）。基于用户唯一标识整合用户和订单数据，构建用户-订单关联表。提取最近30天的数据用于后续计算。 3、数据计算： （1）基于最近30天的早晚高峰时段数据（早高峰7-9点，晚高峰17-19点），利用count函数计算每个用户的最近30天高峰时段骑行次数，进一步计算每个用户最近30天高峰时段每周平均骑行次数=最近30天高峰时段骑行次数/4。 （2）基于最近30天的数据，利用sum函数计算每个用户的最近30天骑行总时长，进一步计算每个用户最近30天每周平均骑行时长=最近30天骑行总时长/4。 （3）按照用户分类模型对用户进行重点标记并进行分类：通勤用户（最近30天高峰时段每周平均骑行次数>3）、休闲用户（最近30天每周平均骑行时长＞120）（上述两类用户均为“重点用户”） （4）基于最近7天的数据，利用count函数计算每个用户最近7天的骑行次数 （5）基于最近30天的数据，利用count函数计算每个用户最近30天的骑行次数，进一步计算最近30天每周平均骑行次数=最近30天骑行次数/4 （6）计算每个用户的活跃系数=最近7天骑行次数/最近30天每周平均骑行次数 4、数据应用：结果数据接入平台，定期对用户活跃度管理数据进行计算，当用户为重点用户且活跃系数低于阈值时触发用户挽回营销方案。This solution analyzes behavioral data such as users' riding duration and riding frequency to identify user riding types, and monitors changes in user activity in real time. When a user's activity declines significantly, the system automatically triggers targeted preferential promotions to improve user retention and repurchase rates. It addresses the ineffectiveness of the "one-size-fits-all" preferential strategy in traditional user operations, enabling precision user retention and optimizing the allocation of marketing resources. 1. Data Source: User data (real-name users) and user riding order data are collected from the operation platform system of Chuangxiang Travel Shared Electric Scooters under Dongyang Chuangxiang Automobile Service Co., Ltd. 2. Data Processing: Clean the collected data to remove abnormal order data (e.g., orders with riding duration less than 1 minute or longer than 60 minutes). Integrate user and order data based on unique user identifiers to construct a user-order association table. Extract data from the most recent 30 days for subsequent calculations. 3. Data Calculation: (1) Based on the morning and evening peak hour data of the most recent 30 days (morning peak: 7-9 AM, evening peak: 17-19 PM), use the COUNT function to calculate the total number of peak-hour rides per user in the most recent 30 days. Further calculate the weekly average peak-hour riding times per user in the most recent 30 days = total peak-hour riding times of the user in the most recent 30 days / 4. (2) Based on the data of the most recent 30 days, use the SUM function to calculate the total riding duration per user in the most recent 30 days. Further calculate the weekly average riding duration per user in the most recent 30 days = total riding duration of the user in the most recent 30 days / 4. (3) Tag and classify users with key labels based on the user classification model: Commuter users (weekly average peak-hour riding times > 3 in the most recent 30 days), leisure users (weekly average riding duration > 120 in the most recent 30 days). Both of the above two types of users are classified as "key users". (4) Based on the data of the most recent 7 days, use the COUNT function to calculate the total number of rides per user in the most recent 7 days. (5) Based on the data of the most recent 30 days, use the COUNT function to calculate the total number of rides per user in the most recent 30 days. Further calculate the weekly average riding times in the most recent 30 days = total riding times of the user in the most recent 30 days / 4. (6) Calculate each user's activity coefficient = number of rides of the user in the most recent 7 days / weekly average riding times of the user in the most recent 30 days. 4. Data Application: Access the result data to the platform, regularly calculate user activity management data, and trigger a user retention recovery marketing plan when the user is a key user and their activity coefficient falls below the threshold.

搜集汇总

背景与挑战

背景概述

该数据集包含东阳共享电动车用户的骑行行为数据，主要用于分析用户活跃度并实现精准营销。数据规模为3001条，每周更新，包含用户骑行次数、时长、分类及活跃系数等多维度信息。