APP User Behavior Analysis Based On Mobile Internet

In order to further promote the Internet process of enterprises,improve the accuracy,completeness and real-time of business data,enhance the APP user experience,optimize access channels,and improve the refined operation effect,more and more enterprises have studied mobile Internet user behavior analysis as a special topic.Through the analysis of the data obtained from the user behavior monitoring,the enterprise can get a more detailed and clear understanding of the user's behavior habits,so as to find out the problems existing in the links of products,subcontracting channels,user operation,marketing and so on,which will help the enterprise to improve the business conversion rate,achieve precision marketing,and thus enhance the enterprise's revenue.Based on the user characteristics and usage behavior of mobile Internet,this paper carries out descriptive statistical analysis through the analysis of transaction APP of S Securities Company user behavior data,uses Dijkstra algorithm to analyze the shortest path,and uses Logistic regression to cluster users.Based on theoretical and empirical studies,the following conclusions are drawn in this paper:(1)Retention analysis: With the passage of time,the retention rate shows a decreasing trend.The next day's retention rate is lower than the industry average.The seven-day retention rate was significantly lower than the six-day retention rate.(2)Distribution channel analysis: the contribution rate of high-quality channel and low-quality channel to active visitors is quite different.(3)Page analysis: the content of the home page is less attractive and needs to be optimized.Users pay more attention to the market,especially the market of the optional stock.(4)Conversion funnel analysis: the user leaks seriously from "transaction login page" to "transaction-ordinary transaction-purchase" steps.(5)Shortest Path Analysis: The long distance of buying function is not conducive to customers' stock,fund and financial transactions.(6)User clustering analysis: Users with relatively high probability of trading behavior have the following characteristics: platform is i OS,channel is website,operator is other,and page browsing time is longer.For the conclusions drawn from the study,this paper puts forward the following countermeasures and suggestions:(1)By increasing marketing efforts and IT security on the trading day,improve the transformation of visitor transactions.(2)Through differentiated recommendation and product optimization,enhance activity.(3)Launch activities and return visits at key time nodes to improve the retention rate.(4)High-quality channels continue to cooperate to ensure user stability,low-quality channels to identify the reasons to enhance the contribution rate.(5)Optimize the home page and insert the main product into the market page to achieve cross-drainage.(6)Optimize the steps of low conversion rate and follow-up analysis of the missing population to improve the step transformation.(7)Shorten the purchasing function path and improve customer trading experience.(8)Select high-value users according to the results of user clustering,and carry out accurate marketing.