The reluctance to report bad news on troubled software projects: a theoretical model

Abstract. By one recent account, only 26% of software development projects are completed on schedule, within budget, and with the promised functionality. The remaining 74% are troubled in some way: they are either cancelled before the development cycle is completed (28%) or are delivered late, over budget, and with reduced functionality (46%). In many cases, the most cost‐effective solution would be to abort the troubled project early in the cycle, but senior managers are often unaware of the project's problems. Anecdotal evidence and at least one recent study suggest that losses are sometimes increased by the reluctance of organizational members to transmit negative information concerning a project and its status. Thus, although evidence of a failing course of action may exist in the lower ranks of an organization, this information sometimes fails to be communicated up the hierarchy or is substantially distorted in the process. The result is that decision‐makers with the authority to change the direction of the project are unaware of its true status. By modelling this reluctance to transmit negative project status information and improving our understanding of the phenomenon, we believe that prescriptions can be developed eventually to reduce the losses from troubled development projects. In this theory development paper, we examine the reluctance to transmit negative information and develop a theoretical model that explains this phenomenon within a software project context. The model we develop draws on literature from the fields of organizational behaviour and communications, ethics, economics, information systems and psychology, and points the direction for future research.