Using the learning curve to maximize IT productivity: a decision analysis model for timing software upgrades

Information technology managers often feel pressure from stakeholders to continually upgrade their firm's software as soon as the software vendor releases a new version. However, this may not be ideal as the software's life cycle and the organization's learning curves are not necessarily aligned. If the firm upgrades its software too early, it may not achieve productivity gains and may lose investments from prior upgrades. If it upgrades too late, it may miss gaining important productivity gains offered by the new version. We present an approach to timing software upgrades that will maximize organization productivity gains. We use the learning curve as the theoretical basis for our mathematical model.