According to Quest Mobile data, the scale of China's mobile Internet users will remain at nearly 1.16 billion in 2021, and the year-on-year growth rate in May 2021 will show negative growth for the first time, which means that the era of traffic dividends is over and it has entered the stage of stock competition. Players on different tracks of the Internet also have to focus on how to improve the retained value of existing users as the key to breaking through the growth dilemma.
In this context, analysts who are reluctant to become a tool for data retrieval can take the initiative to go deep into the business, mine key indicators that affect user retention through massive data, and give reasonable and feasible suggestions based on business logic.
Next, the author will share some mobile number list of the retention analysis methodologies that I have accumulated in my work, and I will give you a lot of dry goods. Don't go away, the exciting will come soon~ The length of the window period depends on the frequency of key user behaviors. For example, car owners and users generally refuel every 14 days, so 14 days can be used as a window period.
The size of the window period can be determined based on the method of the 75% quantile of the interval between two consecutive key behavior days of the user. For example, select users who have performed key actions (such as placing an order or launching an APP) yesterday as the research object. Analyze the number of days between the last two key actions of these users. If 75% of users are within xx days, then xx days can be used as a window for churn.