The analysis of early-stage entrepreneurship is an important stream in entrepreneurship research. The key challenges in this research stream are the dynamics of the complexity of early development processes. We propose a configurational perspective on the analysis of startup development in early-stage entrepreneurship. We argue that a limited number of configurations of startups (ventures immediately after founding) and new ventures (ventures 3.5 years after founding) can be identified, and that there are distinct development tracks between them. The purpose of this paper is to empirically identify these configurations and to investigate their development tracks in order to provide a new explanation of how startups develop.