Yield increases in wheat (Triticum aestivum L.) under the rice-wheat cropping system in Punjab have slowed down in recent years. Further improvements are likely to occur in smaller increments through fine-tuning crop and soil management. For this, a new SSNM approach was planned and evaluated through on-farm experiments conducted for wheat crop from 2002 to 2005 at 56 farmers’ fields across six rice-wheat production domains of Punjab, India. Field- and season-specific NPK applications were calculated using Quantitative Evaluation of the Fertility of Tropical Soils (QUEFTS) model. Nitrogen applications were fine-tuned based on field-specific monitoring of crop N status using chlorophyll meter while P and K applications were fine-tuned based on their respective input-output balances. Average wheat yield under SSNM increased by 0.5 Mg/ha (13%) compared to the current farmers’ fertilizer practice (FFP). These yield increases were associated with a 12% decrease in the average N rate and almost similar P rates but with larger amounts of fertilizer K as compared to FFP. Average N use efficiencies increased by 30 to 40%, mainly through the use of improved crop- and site-specific N management schemes. Compared to the FFP, N applications under SSNM were more uniform among farms, spread more evenly through the growing season and avoided heavy single applications at early growth stages. As a result, SSNM decreased the overall nitrate content in soil profile compared to the FFP but the effect was site-specific. Similarly, P and K balances became less negative under SSNM than under FFP by about 4 and 14 kg/ha, respectively.