Web site data analysis how to define user churn


has not updated the blog for a long time. This article also writes about "user churn". Active users before the release of the site and the loss of the user of this article on the website of the active users, the loss of loss of customers and new customers are defined here, corrected for the loss of the user name English, generally for the loss of the user commonly used English as "churn user", before the use of wastage, away and lost are not too specification. Later a friend asked to do correlation analysis of loss of the length of time the loss of the user’s choice in the end how much is reasonable, especially after the "real" site analysis of this book, I have mentioned the length of time how to more accurately define the loss inside, may explain the relatively simple, there are still problems in this area friend message feedback, so here with an article to explain.

drain users and return users

The loss of the user defined

please refer to the website of the active users and the loss of users "this article to explain how to reasonably define the user loss time problem, need to introduce a new concept: visit customers. Visit customers here does not refer to the Google Analytics Returning Visitor (above access again with new users, refers to the previously visited web site web site), visit the user here refers to the loss again after the users to access the site, the user has a drain, meet the loss within the time limit without access to log on to the site conditions. But after the re access / login website. Then, according to the number of visits, the user can calculate the rate of return visits, that is:

user return rate = return number of users ÷ the number of lost users × 100%

returns the user’s numerical size, indirectly verifying the reasonableness of user churn definitions. Under normal circumstances, the user should return rate is relatively low, from the business perspective, if the definition of loss is reasonable, so it is difficult to make those websites for you have lost interest in the user to access your site. In general, users of the site visit rate should be below 10% in value at around 5% is more reasonable, for mature website users return rate will be slightly higher, while the new website return rate is usually lower, especially products like this kind of mobile phone APP users easy loss.

loss period and return rate of users

The length and duration of

churn churn rate is inversely proportional to the return, we use the definition of user loss when continuous access / login longer after this batch of users visit the site the loss probability will be lower, and with the increase of the loss term defined by the user, the return rate is decreasing, and tends to 0. Then, if you choose the appropriate length of the drain period, we can set a different loss period length >

Leave a Reply

Your email address will not be published. Required fields are marked *