Abstract In the rainfed mid-hill region of Nepal, most fields receive 2–3 t ha−1 of organic compost application every year. Despite efficient recovery and use of organics in the mixed crop-animal systems that predominant in the mid-hills, depleted soil fertility is widely understood to be a significant constraint to crop productivity, with most farmers achieving maize grain yields below 2 t ha−1. Increased use of fertilizer may arrest and even reverse long-term soil quality degradation, but few farmers in the mid-hills use them at present and existing recommendations are insufficiently responsive to site, varietal, and management factors that influence the productivity and profitability of increased fertilizer use. Moreover, policy makers and development practitioners often hold the perception that returns to fertilizer use in the mid-hills are too low to merit investment. In this study, on-farm experiments were conducted at 16 sites in the Palpa district, Nepal to assess the responsiveness of a maize hybrid (DKC 9081) and an ‘improved’ open-pollinated maize variety (‘OPV’, Manakamana-3) to four nitrogen (N) rates, i.e., 0, 60, 120 and 180 kg ha−1, with each N rate response evaluated at 30:30 and 60:60 kg ha−1 rates of phosphorus (P2O5) and potassium (K2O), respectively. With sound agronomy and high rates of fertilizer (180:60:60 kg N:P2O5:K2O ha−1), grain yields observed in the field experiments exceeded 8 t ha−1 with hybrids and 6 t ha−1 with OPV. Yield levels were lower for OPV than hybrid at every level of applied N, but both genotypes responded linearly to N with partial factor productivity for N (PFPN) ranging from 14 to 19 for OPV versus 26–30 for hybrid, with improved N efficiencies obtained when P and K rates were significantly higher. Averaged across phosphorus (P) and potassium (K) levels, a $ 1 incremental investment in fertilizer increased the gross margin (GM) by $ 1.70 ha−1 in OPV and by $ 1.83 ha−1 in the hybrid. For the full response of N, requires higher rate of P2O5:K2O and vice-versa and full response to P2O5:K2O does not occur if N is absent. These results suggest that, i) degraded soils in the mid-hills of Nepal respond favorably to macronutrient fertilizers – even at high rates, ii) balanced fertilization is necessary to optimize returns on investments in N but must be weighed against additional costs, iii) OPVs benefit from investments in fertilizer, albeit at a PFPN that is 36–47% lower than for hybrids, and, consequently iv) hybrids are an effective mechanism for achieving a higher return on fertilizer investments, even when modest rates are applied. To extend these findings across years and sites in the mid-hills, crop growth simulations using the CERES-maize model (DSSAT) were conducted for 11 districts with historical weather and representative soils data. Average simulated (hybrid) maize yields with high fertilizer rate (180:60:60 kg N:P2O5:K2O ha−1) ranged from 3.9 t ha−1 to 7.5 t ha−1 across districts, indicating a high disparity in attainable yield potential. By using these values to estimate district-specific attainable yield targets, recommended N fertilizer rates vary between 65 and 208 kg N ha−1, highlighting the importance of developing domain-specific recommendations. Simulations also suggest the potential utility of using weather forecasts in tandem with site and planting date information to adjust fertilizer recommendations on a seasonal basis.