Recall strategy generation method based on low-steady-state user identification and early warning

Abstract

The invention relates to a recall strategy generation method based on low-steady-state user identification and early warning. The invention provides a method for obtaining user behavior characteristics. The user behavior characteristics are used for describing a user behavior method. The method comprises the following steps: acquiring user behavior characteristics, wherein the user behavior characteristics are used for describing user behaviors; marking lost users according to the user behavior characteristics, and obtaining the number of the lost users; judging whether the number of the lost users is excessive or not, if yes, generating lost user portraits, wherein the lost user portraits are used for expressing lost user features of a lost user group; and obtaining the lost user features, and matching and triggering a corresponding recall strategy according to the lost user features. According to the technical scheme of the invention, before the transaction of the user occurs, the low-stability customer can be found in advance, and the targeted maintenance work can be done for the unstable customer earlier, so that the enterprise is helped to improve the user stickiness, prevent customer loss in advance and improve the operation efficiency, and meanwhile, the customer value is kept.

Description

technical field [0001] The invention belongs to the field of computer data processing, and in particular relates to a method for generating a recall strategy based on low-steady-state user identification and early warning. Background technique [0002] For Internet products, research on lost users is a relatively common and important task in the process of product operation, especially in the operation of new products or immature products. The operation side needs to conduct research on lost users to understand the reasons for user loss, and formulate corresponding strategies to prevent loss, recall lost users, and optimize existing user experience. [0003] In the past, most of the scenarios of lost user research occurred after the product operator discovered the obvious problem of user loss in the operation process before starting to do the research on lost users. Through traditional qualitative in-depth interviews and quantitative questionnaire surveys, the portraits of l...

AI Technical Summary

Problems solved by technology

[0004] Of course, there are also relatively new studies that propose that the characteristic data of lost users can be automatically obtained through technology, that is, according to the preset loss value formula, calculate the proportion of lost user characteristics in the total user characteristics, and calculate the sum of the proportion of each lost user characteristic to automatically obtain The cause of user loss is followed by a recall strategy. However, the reason for the loss obtained in this way is more of the user behavior characteristic values ​​such as user usage time, usage time, etc. The further reasons behind the loss are limited, and the recall strategy is also less effective. Low

[0005] For the existing operation methods, whether it is through the traditional research method or the newer method of automatically obtaining the cause of loss, there are disadvantages. The former has obvious hysteresis effects and cannot automatically warn of loss problems, while the latter can provide early warning. However, the acquired reasons for churn are mostly due to user behavior characteristics such as user usage time and duration of use, etc. The mining of further attitude reasons behind churn is relatively limited, but often in practice, different users have large differences in demand, and existing behaviors The demand reflected by the data may only be the tip of the iceberg of their needs. Due to the lack of precision in the real cause of loss, the recall strategy formulated is also less effective

Images