This paper presents a process that enables real-time correctness of an application to be considered throughout the development and maintenance process. This allows temporal correctness to be ‘designed-in’ rather than attempting to ‘test-in’ timing performance at the end of development. This avoids the costly problems that can arise when timing faults are found late on in testing or, worse still, after deployment. Before presenting this process we must first examine what we mean by real-time and temporal correctness.