We look at the diffusion of seed technology among Ugandan farmers. We present a target-input model to conceptualize the adoption decisions of a new technology in which the best use of inputs is unknown. In this framework, there is path dependency in the adoption process. We show that the group of innovators is well-defined but too small to overcome the system’s inertia. We find little evidence that seed policy reforms implemented in Uganda in the past 20 years boosted agricultural productivity, largely on account of a lack of local experimentation and inadequate use and diffusion of new seed varieties.