Project costs tend to be higher than depicted by budgets. Irrespective of the quality of the project management, this indicates that the budgets are not established in a way which is consistent with how projects are run. Good budgets should on average lead to zero cost overruns. The purpose of this paper is to discuss how project costs (if they are based on models) can be better estimated by taking into account how projects are actually run, in particular, how managerial flexibility adds value to a project. The purpose is not to advocate better ways of running a project, but to obtain better estimates of what projects cost, given the way they are run.