How to conduct user behavior analysis and improve user stickiness?

Retention is considered to be an advanced indicator for measuring product health. In addition to "retention", what we most often talk about is "stickiness", but how to quantify it? Today we share an analysis model - stickiness analysis, which will help you improve your emotional understanding of product stickiness to rational understanding, and help you gain a deep understanding and apply it to business scenarios.

1. Deeply understand retention

For most products, we use retention to evaluate the health of the product as a whole. You can also understand that retention is calculated under the dimension of "how many people use it on a certain day". It counts people from the same group of people over a period of time and calculates the percentage of returning users in this group every day. Taking new user retention as an example, among the new users on a certain day or a period of time, how many people are still using the product on the second day (next-day retention), how many people are still using the product after two days (two-day retention), and how many people are still using the product after seven days (weekly retention). We usually use this to judge the product's ability to retain users and the value of users.

Figure 1: 7-day retention

2. Stickiness: Scientifically evaluate product retention from the user's perspective

From the perspective of refined operations , you may have had such questions: among the user groups that are active at a certain time:

How many users come every 7 days?

How many users did not come for any of the past 6 days?

How many users visited the site for 7 consecutive days?

Among the users who came on the 30th day, how many had not visited in the previous 29 days?

How many users have continuous access?

How many users come 2-3 days a week?

What percentage do they each account for?

If you want to evaluate the health of a product as a whole, we think you may also need to know: "How many days a person uses it", which is a dimension that many products have not been able to measure: stickiness. Because from stickiness you can know: how many days a month a user uses a product, how many use it for more than 1 day, how many use it for more than 7 days . You can also expand it to the weekly dimension, how many use it for more than 2 days a week, how many use it for more than 5 days a week? This is used to comprehensively evaluate the health of the product.

When we visualize this model, as shown in the figure below, select "Any behavior" and view it by week, which is the average distribution of the number of days users use the product per week.

Figure 2: Stickiness analysis of arbitrary behavior

As shown in the figure above, we can see the proportion of users who use the product 2 days, 3 days or more per week on average among all users who have used the product in the past four weeks.

Of course, you can also evaluate the stickiness of a certain function. For example, we choose "Start Sign-in" to analyze the stickiness of the newly launched community function:

Figure 3: Stickiness analysis of the "Start Sign-in" module

Note: When calculating the proportion of people on each day, we will use the number of people who triggered the event during the selected time period as the base number (the first day is 100%). For example, the number of active users in the past four weeks is 200, and the number of people who have triggered "start signing in" is 100. Among them, 20 people have triggered "start signing in" for more than 2 days in a week. Then in the stickiness analysis, the proportion of people who have "started signing in" for more than 2 days is: 20 / 100 = 20%. We do not use the number of active users as the base number. If you want to see the proportion of people who have used a certain function among all active users, you can use the "Active Ratio" function in "Events" to achieve this.

3. Example scenarios

Take our client, xx Financial Services, as an example. As an Internet financial service platform that has been operating for 4 years, the operating idea is the same as all Internet financial products . On the one hand, it needs to continuously strengthen users' trust in the products; on the other hand, by improving the points system /building a shopping mall and other means, it continuously opens up more scenarios for users to interact with products, thereby improving user retention and stickiness.

In addition, we may also need to compare the stickiness of two different user groups. For example, we want to understand how the " invested users" rely on "checking the stock market" compared to the "non-invested users":

Figure 4: Comparison of stickiness of different user groups for "checking the stock market"

(The data is anonymized)

As shown in the figure above, we found that compared with non-investing users, users who have invested pay more attention to the dynamics of the stock market and have a greater stickiness to the stock market function module.

Through stickiness analysis, you can understand how well a product or a certain function retains users. In addition to the commonly used retention indicators, stickiness allows you to understand from more dimensions how users use the product, which functions are liked by users, and what are the differences in how different users use the same function. This helps you evaluate products and functions more scientifically and formulate retention strategies more effectively.

This article was compiled and published by @朱葛君 (Qinggua Media). Please indicate the author information and source when reprinting!

Product promotion services: APP promotion services, advertising platform, Longyou Games