Based on the data of 41 cities in the Yangtze River Delta from 2005 to 2017, the paper uses a modified gravity model to construct a weighted and directed commercial circulation related network, and uses the social network analysis, OLS regression, and QAP regression to research the commercial circulation network's structural characteristics, effect, and influence factors empirically. The results show that, firstly, the nodes in the Yangtze River Delta commercial circulation network are getting closer and more connected, but obvious redundant connections and hierarchical characteristics still exist. Secondly, the different influence of each node in the network is shrinking; the provincial capital city still occupies the leading position of the network center. Thirdly, the eigenvectors centrality will significantly strengthen the linkage strength of each node, the intermediary centrality significantly strengthens the network polarization effect. Finally, the inter-urban differences in the aspects of income levels, local protectionism, financial development, rationalization of industrial structure, Internet development, logistics industry development, and trade openness will affect the evolution of the network significantly. Especially, financial development, Internet information infrastructure and logistics industry play a key role among them.