How to build a user operation system?

However, with the rapid expansion of product scale, facing more and more users, more and more diverse user needs and user scenarios, systematic operation strategies have become an important tool to improve work efficiency and effectiveness. Before specific user operations are carried out, the premise of systematic operations is to establish data indicators for user operations.

Focusing on the three core goals of user operations mentioned in the previous article, when building a user operation system, you should follow the following steps:

• Develop corresponding user grouping and stratification models and indicators (such as ARGO model and RFM model application) according to the goals of different stages;
• Formulate corresponding operation strategies through data analysis, including operation cycle and push method;
• Establish evaluation criteria for data effectiveness and iterate strategies based on operational data.

In the next few articles, we will explain the above steps one by one. In the first part, we will explain how to develop a detailed segmentation and stratification model and indicators for registered users.

Registered users are grouped and stratified, as the name suggests, to simply stratify users and mark them with various tags.

There are a thousand Hamlets for a thousand readers. Although they all use the same product, there are various differences in the reasons why users use the product and the needs they satisfy. Perhaps user A is attracted by the brand sentiment, user B is attracted by the high cost-effectiveness of the product, and user C is attracted by the good service of the product. If we don’t group and stratify users, how can we provide services tailored to their different needs?

Therefore, in the process of user operation, the role of user segmentation and stratification is very obvious. It helps us divide users into various levels and groups, and then we formulate more accurate and targeted operation strategies based on the differences between each level and group. There are several concepts that need to be clarified here, namely "user portrait", "user grouping" and "user stratification". In order to maintain the accuracy of the concepts, we will briefly explain them here:

User portrait: generally includes the user's human attributes, such as identity characteristics, behavioral characteristics, consumption characteristics, psychological characteristics, interests and hobbies, and channel attributes. The specific contents are as follows:

User stratification: Generally speaking, we combine the user's status on the product as the basis for user stratification. For example, our most common RFM model relies on the user's most recent purchase time, consumption frequency and consumption amount, and stratifies users by measuring user value and user profitability.

For example, we use the RFM model to divide users into eight groups, ranging from important value users with high consumption frequency, high consumption amount, and recent consumption, to general retention users with low consumption frequency, low consumption amount, and no consumption for a long time.

RFM Model

User segmentation: Compared with user stratification, user segmentation focuses more on user behavior performance, such as the ARGO growth model of intelligent user operation launched by iResearch Ark.

For example, high consumption frequency + low consumption amount, and low consumption frequency + high consumption amount, these two types of users may both be high-consumption users in a certain sense, but their behavioral characteristics will be very different, and the corresponding operating strategies will also be different.

In fact, it can also be understood that user grouping is a further refined division of user stratification. The relevant indicators of the ARGO growth model have an obvious progressive relationship, while the RFM model is independent of each other.

Now that the above keywords have been explained clearly, let’s move on to the detailed segmentation and stratification of registered users.

Because user segmentation and stratification may be diverse in different industries, user segmentation and stratification also change differently at different stages of product development, and user segmentation and stratification require both qualitative and quantitative analysis. Therefore, we can follow the following two principles to help us better complete user grouping and stratification:

(1) Fine grouping and stratification, following the MECE principle

In the process of fine-grained segmentation and stratification of registered users, we follow the MECE analysis method proposed by McKinsey. This not only helps user operations find all key factors that affect the expected goals and find all possible solutions, but also helps to sort and analyze users, problems or solutions and find satisfactory solutions from them.

(MECE analysis, in Chinese, means "mutually independent and completely exhaustive". That is, for a major issue, it can be classified without overlap or omission, and it can effectively grasp the core of the problem and become an effective way to solve the problem. - Baidu Encyclopedia)

That is, on the basis of establishing the main goal, we then break it down layer by layer until all the sub-goals are found. By decomposing the goals layer by layer, we can analyze the relationship between the user's key behaviors and the goals.

(2) Clarify the stage goals to make grouping and stratification easier to understand

Just as users have a life cycle, product and user operations also have clear phased goals and strategies. The product's life cycle stage is different and the requirements for user operations are also different.

Author: Yi Analysys

Source: Analysys