Industry classification is important for industry analysis and competitive intelligence. However, existing schemes and methods are limited by the small number of industry categories and the lagged information of firms’ business. In this paper, we propose a novel industry classification method by constructing the labor mobility network from the LinkedIn profiles. We also propose a hierarchical extension of the community detection algorithm to better discover latent industry clusters on the constructed network. The evaluation conducted on real-world datasets shows that our method outperforms the best existing industry classification scheme and the state-of-the-art method and improves their explanatory power by 8.31% and 3.97% respectively. Moreover, our method is effective in earlier revealing firms’ action of entering new industries.