Entrepreneurial strategies, conceptualised as theories of value, rely on novel and actionable causal frameworks to guide decision making under uncertainty. This paper explores the principal properties of good theories of value—novelty, simplicity, falsifiability, and generalisability—focusing on the implicit trade-off between novelty and simplicity to identify the optimal complexity of such theories. We introduce a formal model to demonstrate how entrepreneurs construct parsimonious theories, balancing the benefits of additional causal attributes against the cognitive and resource costs of complexity. The model predicts that more novel problems require more complex theories, while higher cognitive ability enables entrepreneurs to handle greater complexity effectively. To test these predictions, we propose a randomised controlled experiment to observe theory development in varying novelty conditions leveraging a bespoke online platform developed specifically for theory creation using Bayesian networks. This study contributes to the theory-based view by micro-founding parsimony as a critical property of entrepreneurial theories, enhancing both academic understanding and practical guidance for entrepreneurs.
