We consider managerial decision-making regarding the evolution of knowledge in a three-stage new product development project. The manager invests in knowledge development activities (such as prototyping, pilot line testing, ramp-up experiments) at each stage throughout the project. The links between development activities at different stages are captured by recognizing that, as a result of knowledge transfer, the ability of the recipient team to generate new knowledge is enhanced. Over time as the levels of knowledge increase, product features and process characteristics improve. The performance of the new product in the marketplace, which drives net revenue, reflects the levels of knowledge attained at each stage of the project at the product launch time. The objective is to maximize the net revenue earned when the product is released to the marketplace less development costs. We show that the rate of each development activity follows an entirely different dynamic strategy during the project. In the first stage, development activities follow a front-loading strategy; in the second stage, development activities follow a moderate delay strategy, and in the third stage development activities follow an extreme delay strategy.