This study explores how exploration strategies, sequential or parallel, impact commercialization outcomes of deeptech ventures, offering insights into the effects of automation and lifecycle acceleration on firm performance. Lean methods promote early commercial exploration and parallel experimentation to address entrepreneurial uncertainty. Deeptech firms operate in environments characterized by multiple uncertainties, including technical, commercial, and regulatory risks. Using a novel longitudinal dataset of 13,790 firms gathered through machine learning/ NLP and other techniques, and leveraging a difference-in-difference identification strategy based on an exogenous technology shock, I find that parallel exploration improves short-term financial performance by mitigating information asymmetries with resource providers. In contrast, sequential exploration improves organizational learning and is likely to yield superior long-run innovation outcomes. These findings highlight the managerial trade-offs between addressing information asymmetry and enhancing organizational learning for boundedly rational entrepreneurs. This research contributes to literature on experimentation and learning while also informing emerging work on the effects of general-purpose technologies on firms, particularly at the downstream stage of commercialization. It also sheds light on managerial implications in terms of the boundary conditions for employing lean methods with science and technology ventures.
