2500 words user retention analysis

Faced with the current situation of difficulty in increasing traffic, companies should think about how to retain users , implement refined operations, and drive subsequent growth with lower costs for existing users. So, in what scenarios is retention analysis generally applicable? How should it be applied?

In this user-centric Internet world, let's discover more interesting points together...

1. Retention issues faced by Internet companies

1. Traffic dividends have peaked and the cost of attracting new users is high

In today's world where homogeneity is extremely easy, competition for traffic is extremely fierce. The importance of improving user retention is self-evident, and the cost of acquiring existing users is much lower than the cost of attracting new users.

2. New users are more likely to churn

Companies invest in advertising and organize events on a large scale, but the high expenditures cannot lock in new users, and their long-term sustainable development is somewhat weak. They only care about immediate interests and only treat the symptoms but not the root cause.

2. What is retention analysis?

Retention rate: The proportion of users on a certain day who still launch the app on the Nth day. Retention analysis is to analyze the activity of users over time. Acquiring users is only the first step; retaining users is the ultimate goal of all products.

It can be understood as the process of converting initial wavering users into loyal and stable users. The higher the retention rate, the stronger the user's dependence on the product.

It can be divided into three stages:

1. Initial stage: New users have just registered and user retention rate drops rapidly, so users need to quickly feel the core value of the product.
2. Medium term: New users settle down and become active users. At this time, it is necessary to analyze active retention, strengthen core functions, and cultivate user habits for product usage.
3. Later stage: think about the core value of the product and carry out product iteration and optimization.

1) Divide from the time dimension

Common ones are: next-day retention, 3-day retention, 7-day retention, 30-day retention, weekly retention, and monthly retention.

2) Segmentation from the user dimension

Common ones include: new user retention and active retention.

The diagram is as follows:

3. Commonly used calibers for retention analysis

1. Take new user retention as an example

• Next-day retention rate = (number of new users who logged in on the second day after registration) / number of new users on that day;
• 3-day retention rate = (number of new users on a certain day who logged in on the third day after registration) / number of new users on that day;
• 7-day retention rate = (number of new users who logged in on the 7th day after registration) / number of new users on that day;
• 30-day retention rate = (number of new users on a certain day who logged in on the 30th day after registration) / number of new users on that day;
• Retention rate after 1 week = (number of new users in a certain week who logged in in the second week of registration) / number of new users in that week;
• Retention rate after 2 weeks = (number of new users in a certain week who logged in in the third week of registration) / number of new users in that week;
• Retention rate after one month = (number of new users in a certain month who logged in in the second month of registration) / number of new users in that month;
• Retention rate after 2 months = (number of new users in a certain month who logged in in the 3rd month of registration) / number of new users in that month.

2. Take active retention as an example

• Next-day retention rate = (number of users who logged in on a certain day and logged in on the second day) / number of users who logged in on that day;
• 3-day retention rate = (number of users who logged in on a certain day and logged in on the 3rd day) / number of users who logged in on that day;
• 7-day retention rate = (number of users who logged in on a certain day and logged in on the 7th day) / number of users who logged in on that day;
• 30-day retention rate = (number of users who logged in on a certain day and logged in on the 30th day) / number of users who logged in on that day;
• Retention rate after 1 week = (number of users who logged in in a certain week and logged in in the second week) / number of users who logged in that week;
• Retention rate after 2 weeks = (number of users who logged in in a certain week and logged in in the third week) / number of users who logged in that week;
• Retention rate after 1 month = (number of users who logged in in a certain month and logged in in the second month) / number of users who logged in in that month;
• Retention rate after 2 months = (number of users who logged in in a certain month and logged in in the third month) / number of users who logged in that month.

IV. Applicable Scenarios for Retention Analysis

1. Daily retention rate

• Quickly determine whether the product meets market demand, such as whether novices are satisfied with the product's UI design, function settings, and novice guidance, and whether adjustments are needed.
• Quickly determine user stickiness, such as whether users are more susceptible to promotional activities, etc.

2. Weekly retention rate

• To judge user loyalty, users have basically had a complete experience of the product at this point. After going through the entire process, users who continue to visit can be judged as potential loyal users.
• Analyze the reasons why users visit again, find out the points of the product that can best retain users, and refer to these points consistently, expand the application to more users, and encourage more users to stay.

3. Monthly Retention Rate

Evaluate the effects of iteration and optimization. Cut off product features with low retention rates and perform iterative optimization.

5. Possible reasons for the decline in retention analysis

1. Decreased retention of new users

• New users did not quickly experience the core value of the product;
• The novice guidance module provides good experience;
• Most of the new users are freeloaders;
• Interface UI design affects the user experience;
• Poor product function experience;
• …

2. Decline in retention of old users

• Product iteration functions lead to a worse user experience;
• The product iteration cycle is long, and users lose their sense of freshness;
• Affected by competitors;
• Not encouraging users to form habits with the product;
• The discount for continuous clocking in and signing in to get red envelopes is relatively small, so there is no point in persisting;
• More advertising push;
• Customer service is slow to respond and provides poor service;
• No relevant push;
• There are many product bugs;
• Affected by promotional activities;
• …

6. Retention Analysis Method

Among them, product function analysis: Purpose: to find out the most valuable and least valuable functions for retention, so as to facilitate later iterative optimization.

• Excellent functionality: It is recommended to focus on optimizing the user experience.
• Public functions: This is of utmost importance. It is recommended to reflect on the long-term value and practicality of this function.
• Niche features: It is recommended to keep this feature, but there is no need to invest too much effort in it.
• Weak function: It is recommended to consider whether to remove it.

7. Case Study

1. Case 1

This picture was processed by me on PPT, and two days were selected for comparison.

1) Analysis

The retention rate of new users registered on May 1, 2021 tends to be stable on the 7th day of registration, at which time the retention rate is 60%; the retention rate of new users registered on May 2, 2021 tends to be stable on the 7th day of registration, at which time the retention rate is 20%; the stable retention rate of users registered on the 2nd is worse than that on the 1st.

2) Improvement ideas

The retention rate when it tends to be stable should be increased as much as possible, that is, the stable line should be raised as high as possible.

2. Case 2

The data is purely personal fiction. It is recommended to expand the date in actual analysis. This chart focuses on analyzing the analysis method.

Retention rate of this table: (number of new users who logged in on day N) / number of new users on that day

Take the retention of new users on August 1 as an example.

• Novice exploration period: Users attracted simply by large discounts will be lost sooner or later, and the product value will not meet user expectations.
• Habit formation stage: Product functions and practicality do not encourage users to form usage habits.
• Active user period: truly loyal users who stay.

Analysis

• New user retention rate dropped by 60%: users failed to quickly discover the value of the product.
• The overall retention rate tended to stabilize on the 10th day, and the retention rate stabilized at around 11%, indicating that only about 11% of the new users on August 1 became loyal users.
• The retention rate of 3rd and 7th retentions increased (note: the retention rate does not show a continuous decline). To further identify the reasons, were there any promotional activities on August 3rd and August 7th?

3. Case 3

Analysis

The table takes the second retention rate (71%) of users who registered on August 6 as the starting point, and the seventh retention rate (34%) of users who registered on August 1 as the ending point. The two form a diagonal line. Comparing the data vertically, the retention rates of the color parts are relatively high.

First of all, we need to confirm whether the operation took any action on August 7? For example: Was there any promotion or other special event on that day? Because August 7th corresponds exactly to the second stay on August 6th, the third stay on August 5th... and the seventh stay on August 1st.

In the table, the second retention rate on August 9 is 20%, which is much lower than that on other days, and the subsequent retention rate is also lower than that on other days. Be wary of freeloaders.

Author: Tableau from Beginner to Mastery

Source: Tableau from Beginner to Master