Specialization vs. Generalization : Analyzing Skill Transferability for Predicting Career Trajectories in High-Tech

For decades, career transition counseling and research have focused on myopic insights on the next job versus understanding optimal career path. While the research mind-set was well placed to understand the evolution of skills to suggest possible terminal job outcomes, a lack of relevant data and computation resources posed challenges in developing valuable career transition insights. Furthermore, because of the data limitations, the extant research focused on the role of generalizable and specialized skills in highly heterogeneous career transitions. Technology, big data, and analytics have now enabled organization and management science research to investigate these questions. This paper analyzes 67,000 career profiles for entrepreneurs, executives, and senior managers from a popular social networking platform to predict career outcomes for these professionals. We built career trajectories as network graphs that provide insights on career paths of entrepreneurs, executives, and senior managers. In these graphs, each node is a job consisting of an industry-level and seniority-level classification. Focusing on the high-tech industry, we found that individuals with certain transferable skills—notably technical, management, mixed, or boundary-spanning experience—can successfully capitalize on these skills when transitioning across industries. During the process we determined common patterns in career trajectories for entrepreneurs, executives, and persons in senior-level roles within firms. Furthermore, we tested the validity of these insights by developing and refining a learning model for predicting individual career transitions of 10,000 individuals and achieved 48% accuracy for 204 job choices. Managerially, these insights can help in recruitment (candidate identification), career planning, and job search