lead: funnel model and attribution model not only can be used in data operation work, but also for the overall business process of combing and clear. This can also be said to be one of the essential skills in the work.
data operation classmates will have such confusion: data operation difficult? I am not good at math do? Is not also need to learn mathematical modeling? I should see what book learning? Including meow started working in your face a confused, do not know from business data how to start. Today we have the common product data together with the smooth smooth work analysis method, answer the problem of "how to do".
I have the couple said, data can draw user portrait and behavior trajectory, can monitor the product transformation and development, can assess the transverse channel effect, all of which relates to the cooperative work of many departments and products, market and technology. Data management is a great and a lot of hands.
today is the first to introduce the funnel model: it can be widely used in traffic monitoring, product target conversion and other daily data operations. The reason is called the funnel, because the user (or traffic) from a centralized point of entry (which can be set according to the business needs of their own), the product itself may be set to complete the operation of the process.
we have to do is to monitor each process in accordance with the operation of the user level conversion, find the optimization point of each level; not in accordance with the operation process the user to draw the transformation path of them, can be found to enhance the user experience, to shorten the path space. The use of funnel model is a typical case of the transformation of the electricity supplier website, the user in the purchase of goods will be in accordance with the pre design of a single purchase process, the final payment. Although these data I made up (you call me) but if not the entire business process of carding, would not have this funnel model output, let alone to find every step appears churn problem.
some time to do some competitive analysis, for the transformation of industries with similar data to know what to do. To reduce the loss of customers is our goal as much as possible, but if you can do no less than the industry average and limited resources, reduce the conversion funnel loss of customers need to be placed at a lower priority.
there are some of the more classic funnel transformation model for user registration process is: we need to know how many users click the register button (funnel start), the number of users completed the information form (the number of users, many users give up fill) click send verification code button (verification code, arrival rate) successfully completed the registration number. If the number of registered users in the course of the day of operation is found to fluctuate