Monitoring and evaluation (M&E) is an extremely complex, multidisciplinary and skill intensive endeavor. Government-wide M&E is even more so because it requires detailed knowledge across and within sectors, as well as of interactions among planning, budgeting, and implementation functions in the public sector. The situation is complicated even further when the machinery of government is decentralized, with powers and functions distributed across three spheres of government. This paper outlines the process of implementing a government-wide monitoring and evaluation (GWM&E) system in South Africa. The first section sketches the context that created the impetus for establishing such a system. This context is clearly shaping the evolution of the system and influencing its longer-term sustainability. The second section outlines the various stages of conceptualizing and implementing the GWM&E system, which is currently very much a 'work in progress.' The third section reviews international experiences for lessons learned, which may also be germane to the South African context, noting similarities and differences in approach. Some of the critical implementation factors relate to the role of political leadership and championing of M&E, incentives for promoting usage of M&E findings, dealing with information and data constraints, capacity building, "ownership" of the M&E system by line ministries and other agencies, and managing the challenges of change. The fourth section examines a range of challenges and difficulties encountered in South Africa. The final section reflects on lessons distilled from the South African experience to date.