We consider Just-in-Time (JIT) to be an overall organizational phenomenon. Accordingly, we developed and tested a model that includes both JIT practices and the infrastructure practices hypothesized to provide an environment in which JIT practices perform more effectively. Canonical correlation analysis was used to test five hypotheses. The results indicated that: (1) there was not a significant relationship between the use of JIT practices, alone, and manufacturing performance; (2) there was a very strong relationship between JIT practices and infra-structure practices; (3) the combination of JIT management and infrastructure practice was related to manufacturing performance; (4) infrastructure, by itself, is sufficient to explain manufacturing performance; and (5) manufacturing performance was related to competitive advantage. These findings provide support for the notion that JIT is an overall organizational phenomenon, rather than limited to strictly shop floor practices, and that at least part of its effect on manufacturing performance may be through providing a set of improvement targets and discipline for the entire organization. In addition, the analysis highlights the areas of infrastructure practice most relevant for future research.