Event analysis: View the usage of function points; Funnel analysis: monitor the registration process, registration page - initiate registration - complete registration. (2) After monitoring, how to find product abnormalities? Product overview: For example, the registration rate whatsapp Database has decreased by 20% compared with the same period last week; Page analysis: The number of visits to the registration page increased slightly by 3% compared with the same period last week; Event analysis: The per capita usage of the verification code refresh button has whatsapp Database increased by 100%; Funnel analysis: The registration funnel, the registration page - the conversion rate of the link to initiate registration decreased by 30% compared with the same period last week.
After an abnormality is found, how to analyze the cause of the abnormality? Through the dimension of mobile phone model X version, it is found whatsapp Database that all registration rates have dropped; Go to the funnel analysis and find that the registration page-initiating registration link has a serious loss; Monitor the event triggering of the registration page, and find that the number of refreshes per capita of the verification code has skyrocketed; Confirmed whatsapp Database with the product operation, in order to prevent malicious registration, the graphic verification of the verification code was updated yesterday. The new graphic verification code is difficult to identify with the naked eye, so users need to refresh the verification code frequently.
(After solving, how to verify the optimization effect? Product Overview: The registration rate rebounded to the level of the same period whatsapp Database last week. 4. Content output personnel: (video as an example) (1) How is the video viewing situation now? Video effect: How many people saw it? How many people clicked? How many people play? How long is the whatsapp Database playback time? How many people watched it? How many people have initiated comments? How many likes? What is the percentage of bad reviews? What is the percentage of forwarding?