Our goal is to establish a science of the individual, grounded in dynamic systems, and focused on the analysis of individual variability. Our argument is that individuals behave, learn, and develop in distinctive ways, showing patterns of variability that are not captured by models based on statistical averages. As such, any meaningful attempt to develop a science of the individual necessarily begins with an account of the individual variability that is pervasive in all aspects of behavior, and at all levels of analysis. Using examples from fields as diverse as education and medicine, we show how starting with individual variability, not statistical averages, helped researchers discover two sources of ordered variability--pathways and contexts--that have implications for theory, research, and practice in multiple disciplines. We conclude by discussing three broad challenges--data, models, and the nature of science--that must be addressed to ensure that the science of the individual reaches its full potential.