This paper draws on the literature in management science, information systems and organization development along with practical experience to propose a process of building models that will be implemented. The steps are: (1) formulation of priors, (2) entry, (3) problem finding, (4) specification of model development criteria, (5) model building, (6) estimation and fitting, (7) tracking, and (8) continuing use. After describing the state of the art of model implementation, each step is discussed. It is hoped that this exposition will encourage comments and lead towards the specification of a methodology for model building that will make real the promise of management science.