Purpose This paper explores the problems of managing superprojects and identifies how a different approach to controlling them can reduce the incidence of cost and time overruns and benefit shortfalls. Design/methodology/approach Literature review accompanied by conceptual analysis. Findings Project cost and timing overruns and benefit shortfalls are very frequent in superprojects. These problems can be ascribed partly to the way in which they are planned is not taken into account in designing and implementing control systems, particularly the governance processes and the information they have available. Practical implications This article has serious implications for those designing control processes, governance and information management for superprojects. It suggests that if a new approach is taken, fewer superprojects will suffer from cost overruns and benefit shortfalls, because remedial actions will be taken earlier for projects which are experiencing problems, while learning will be fed back to those planning new projects. Social implications There will be saving of public money and reduced deferment of benefits that normally result from failed or delayed projects and reduced allocation of large incremental budgets dedicated to resolving problems. Originality/value The taxonomy of different types of superprojects is original, as is the idea of ambidextrous control, and the diagnosis of failure reasons lying in the nature of control and governance processes and the lack of relevant information available during the control process. Keywords Project planning, superproject, project management, programme management, transformation, bias, control, information management Paper type Conceptual paper.