Value to be determined. The above is just a category email list simple analogy and may differ from the actual figures, but qualitative research shows that the cost of absorbing new users is 5-10 times the cost of maintaining original customers. If you cannot measure it, you cannot improve it. What you can't measure, you can't improve. Reducing the cost category email list of user growth is the goal, but the process and method cannot be ignored. User retention Retention :

Count the proportion of customers who still have category email list customer behavior (such as login and consumption) after a period of time among new customers in a certain period. Retention actually decomposes the user's inhalation and loss from the two dimensions of relative time and absolute time . In order to better describe the changes in user retention category email list in the time dimension, Cohort Analysis is generally used in the industry.

Therefore, after borrowing the e-commerce customer category email list data provided by UCI as a demonstration and using Python to do the necessary data processing, the author presents the following analysis diagram of Cohort Analysis: Absolute time dimension: The calendar time when the user first used the product. For example, March 1, March 12, 2020. Relative time category email list dimension but the process and method cannot .