Entrepreneurial learning, characterized as entrepreneurs adapting their strategies in response to feedback, is considered a crucial process for success. However, the feedback entrepreneurs learn from is ambiguous because it comes from multiple customers who value different aspects of performance. The resulting information consists of mixed signals – both across different performance metrics and across the individuals providing feedback for a given metric. This dissertation focuses on the interpretation problem of adapting to ambiguous feedback. I examine how ambiguity in feedback impacts entrepreneurs’ adaptive responses, and the performance implications of those adaptive responses. The first part of the dissertation explores how entrepreneurs selectively adapt to multidimensional feedback, exploring their nuanced responses to positive and negative feedback. I find that entrepreneurs are more likely to respond to dimensions receiving positive feedback. Furthermore, the extent of feedback ambiguity influences this tendency in two ways: it intensifies when feedback spans multiple dimensions, but diminishes when conflicting positive and negative signals exist within a single dimension. The second part of the dissertation examines the performance consequences of adaptive behaviors. Contrary to the conventional view that selective adaptation to positive feedback reflects biased judgment, I view this behavior as an attempt to build on the firm’s potential competitive advantages. I also find that this selective incorporation of positive feedback is more effective under higher competitive pressure, which necessitates a narrow and differentiated niche, and when a firm is engaging in a differentiation strategy, which creates a buffer to early competition. Empirically, I examine these processes in a global video gaming platform that hosts early-stage games that users provide feedback on over the course of the game’s development. Using natural language processing (NLP) techniques, I compare the contents of the feedback to the game updates to determine which feedback is incorporated into the product, how the behavioral patterns are influenced in the face of ambiguity, and the performance implications of those actions.