CONTEXT Cocoa swollen shoot virus (CSSV) remains the most destructive viral disease affecting cocoa in West Africa, with ecological and economic repercussions. Since the virus is specialized to cocoa and disperse locally through mealybugs vectors, landscape structure could play a crucial role in determining epidemic outcomes. OBJECTIVES This study aimed to assess how landscape structure influences CSSV spread and economic returns, to identify landscape structures that reconcile epidemiological control with profitability. METHODS The spatially and temporally explicit landsepi model was parameterized for the CSSV pathosys tem using literature-derived data and calibrated through Approximate Bayesian Computation based on field observations. The model was then used to compare landscape mosaics varying in cocoa (host) and coffee (non-host) proportions and spatial aggregation, under contrasting disease pressure, economic, and yield scenarios over a 20-year period. RESULTS AND CONCLUSIONS Epidemiologically, long-term cocoa survival improved in fragmented landscapes, with optimal outcomes below 40% cocoa under low disease pressure and below 20% under high pressure. However, these compositions reduced total system income. Economically, cocoa-dominated landscapes remained profitable under low disease pressure. Under high disease pressure, profitability persisted only in the short (4–5 years) to medium term (12–13 years), depending on cocoa–coffee gains differences. Over longer periods, profitability declined unless profits were accumulated, indicating that short-term benefits only partly offset long-term losses. SIGNIFICANCE Collective, landscape-scale diversification emerges as a promising strategy to contain CSSV while sustaining cocoa production. By jointly considering disease pressure and landscape structure, stakeholders can design context specific landscape mosaic that reconcile epidemic control with economic profitability.