Chari, MuraliAlrubaiaan, Omar2025-11-232025https://hdl.handle.net/20.500.14154/77132This dissertation consists of three essays on entrepreneurial strategy and performance. Entrepreneurship is inherently risky, and understanding the factors that shape entrepreneurial outcomes remains a central concern in the field. The three essays highlight complementary themes: strategic adaptation through pivots and the use of advanced AI to detect them, the role of entrepreneurial experience in shaping performance, and the contextual conditions under which immigrant entrepreneurs achieve superior outcomes. The first essay introduces a novel approach to measuring post-launch pivots in startups using Large Language Models (LLMs). By systematically analyzing changes in firm descriptions over time, the study develops an automated method to capture strategic redirection at scale. While pivoting is widely discussed in entrepreneurship, prior research has largely relied on case studies of individual or small groups of ventures. This study advances the literature by leveraging CrunchBase data and prompting LLMs to assess description changes and benchmarking their performance against human raters. Results show that the top-performing LLM outperforms human accuracy (84% versus 79%), demonstrating the promise of LLMs as scalable tools for systematically studying pivots across large samples. Moreover, when human raters were subsequently exposed to the LLM’s assessments, their accuracy increased to 85%, demonstrating the potential of LLMs not only as scalable analytical tools but also as valuable decision-support systems that enhance human judgment. The second essay examines the performance of serial versus novice entrepreneurs, distinguishing between survival and financial outcomes. Using panel data, the study develops and tests hypotheses on differences in venture performance, arguing that survival and financial performance—often treated interchangeably—reflect distinct dimensions of entrepreneurial success. Findings reveal that the first ventures of serial entrepreneurs generate higher financial performance but shorter survival relative to those of novices. Furthermore, subsequent ventures of serial entrepreneurs survive longer than earlier ones, though their financial performance does not significantly improve. These results shed light on the nuanced relationship between entrepreneurial experience and venture outcomes. The third essay investigates the performance of self-employed immigrant entrepreneurs in the United States, asking when and under what conditions they outperform their non-immigrant counterparts. Drawing on longitudinal panel data from the PSID, the study tests four hypotheses, including the moderating roles of family size, wealth, and industry context. Findings show that immigrant entrepreneurs outperform their native-born counterparts on average, with family size strengthening this advantage and wealth diminishing it. Industry dynamics also matter: immigrants excel in blue-collar sectors but are less competitive in white-collar industries, reflecting both constraints and opportunities embedded in immigrant entrepreneurship. Taken together, these three essays contribute to entrepreneurship research by advancing methodological tools for measuring pivots, clarifying the performance implications of entrepreneurial experience, and uncovering the contextual factors that shape immigrant entrepreneurs’ outcomes. Collectively, they provide a multidimensional perspective on entrepreneurial strategy and performance.148en-USEntrepreneurshipInnovationThesis