Non-union is a significant problem in healthcare systems. With limitations to current data collection, we need to devise robust prospective database to accurately estimate the incidence of non-unions. In doing this, we will gain insight into the magnitude of non-unions and allocate resources accordingly through provision of tertiary treatment centres and in research planning. By focusing on early accurate diagnosis, through a standardised pathway we can identify trends and risk factors. The shift towards standardised definitions and data homogeneity will facilitate the generation of predictive models. The models will not only risk stratify patients but highlight a new focus for research. By risk stratifying patients, we can counsel them accordingly and adjust expectations and reduce indirect costs, which may reduce the litigation burden. Management plans can focus on reversible risk factors to enhance outcome of fracture management and patient satisfaction.