Despite the continuous need to enhance the value of public services, information systems projects in the public sector still underperform due to their high velocity of change and strategic ambiguity. Diverse studies have been carried out examining 'project capabilities' to contribute to the efficiency of information systems project management. However, most studies of project capabilities focus heavily on the project execution that aims for successful implementation of systems. This supplier-oriented approach leads to overlooking the significance of a project owner's benefits realisation after implementing the information systems. Moreover, this formulation of project capabilities also does not distinguish between dynamic capabilities for benefits realisation and operational capabilities for project execution. For these reasons, the realisation of a project owner's information systems benefits has still been far from satisfactory, in addition to having a poor rate of project success. In order to address this issue, this thesis develops a more nuanced perspective on project capabilities by distinguishing the dynamic capabilities of owners from the operational capabilities of suppliers and by developing the concept of owner dynamic capabilities. This is followed by analysis of the importance of benefits management as an owner dynamic capability. The aim of this study is to contribute to a deeper understanding of why public sector information systems projects are so challenging and how the project owner's information systems benefits can be accelerated. A content analysis method was adopted, and 10 years of National Audit Office Value-for-Money reports were analysed covering 31 information systems projects. Theoretically the concept of 'owner dynamic capabilities' is introduced, and the experience of UK central government information systems projects is empirically reviewed. The results bring a key implication by showing the significance of benefits management as a distinctive 'project back-end' owner dynamic capability. This thesis makes three main contributions to the current literature in information systems project and benefits management. First, the concept of owner dynamic capability is introduced and theorised based on a unique data set of major public sector information systems projects. Second, the importance of project back-end capability is revealed as a distinctive dynamic capability, which owners require to move their information systems investment from practical completion (the system works as expected) to beneficial use (the system delivers the expected business benefits). Third, the key findings provide the future research agenda for project management disciplines.