Purpose: As part of the institutional reforms and agricultural restructuring in the Democratic Republic of Congo (DRC), this paper provides an assessment of the performance of the agricultural extension system as well as factors explaining it. Method: This paper involves key informants' interviews and surveys of 107 extension organizations and 162 extension agents in randomly selected 156 villages, analyzed using qualitative and logistic regression methods. Findings and Practical Implications: Results show that despite having one of the highest extension agent-to-farmer ratio and a pluralistic extension system, DRC fails to deliver knowledge and technologies to rural areas due to lack of coordination, no unified and clear policy and mandate, lack of funding, aging and low competencies of agents, and lack of mobility and interactions of agents with key factors. This paper complements findings by other studies that number of agents is not a sufficient indication of performance, but an effective system needs to focus on the enabling environment for agents to be motivated to work as mandated. In this paper, enabling conditions that are found to be statistically significant are external funding, enforcement of performance targets, systems of rewards and sanctions, mobility to foster linkages, and skills development. Originality: This paper contributes by: (1) analyzing a cross-section of various organizations and agents to identify factors that explain variations in performance in a statistical and systematic approach; (2) providing insights on how to prioritize investments and options for a fragile state like DRC, with weak infrastructure and institutional capacity and with a long history of neglect for their national extension system; and (3) illustrating how a rich and well-cited conceptual framework can be implemented empirically to provide policy options for a country like DRC.