With the development of technology, the data stored by humans is growing geometrically. Especially in the logistics industry, the rise of online e-commerce has created a huge data flow in the informatized logistics network. How to collect, analyze, and organize this information in time and analyze the meaning of this information from it is a difficult problem. The paper aims to learn the management of logistics systems from the perspective of statistics. This article uses random analysis of 1,000 customers’ logistics records from the logistics enterprise information system, uses mathematical analysis and matrix theory to analyze the correlation among them, and analyzes customer types and shopping. The information on habits, daily consumption patterns, and brand preferences is classified and summarized using mathematical statistics. The experimental results show that the results of the study can well reflect customers’ daily habits and consumption habits. The experimental data show that mining effective and accurate information from massive information can help companies to quickly make decisions, formulate scientific logistics management programs, improve operating efficiency, reduce operating costs, and obtain good benefits.