LLM Analysis of 130K LinkedIn Profiles Reveals Career Mobility Secrets for Planners

Researchers Yan Wang and Su Jeong Jo published a study on arXiv (2605.12618) that leverages large language models to parse and structure data from over 130,000 LinkedIn profiles of U.S. planning alumni. The work applies boundaryless career theory, social capital theory, and spatial opportunity models to uncover what drives career advancement in a rapidly urbanizing landscape. By using LLMs to extract job titles, skills, industries, and network size from unstructured profile text, the study offers a scalable alternative to traditional surveys.

Key findings reveal that planning alumni who adopt boundaryless career patterns—such as moving across sectors or making lateral industry switches—achieve significantly higher upward mobility. While technical competencies help early on, soft skills become more decisive as professionals reach senior stages. Geographic mobility and working in larger, diverse metro areas also correlate with advancement, though the latter provides only modest benefits. Notably, AI-related skills, now common, offer little additional edge. Larger professional networks and organizational engagement consistently predict upward transitions. The study acknowledges limitations, including LinkedIn’s potential underrepresentation of those without profiles and an individual-level focus that omits organizational factors.