Field experiments were performed over 3 yr to examine the impact of insecticide application timing to control soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae), populations and to prevent soybean yield losses. Experiments were conducted in early and late-planted soybean, Glycine max (L.) Merr. Insecticide applications were made based on soybean growth stages. In 2001, applications were made at V1, V3, R2, and R3 growth stages; in 2002 and 2003, applications were made at R2, R3, and R4 stages. Additional treatments consisted of an unsprayed control and a multiple spray treatment that received insecticide applications at 7–10-d intervals. Soybean aphid densities were recorded throughout the growing season, and yields were measured. Soybean aphid populations varied considerably across years and planting dates. In general, late-planted soybean exhibited higher aphid pressure than early planted soybean, and experiments in 2002 had lower aphid numbers than those in 2001 and 2003. The multiple spray treatment significantly increased yield over the control in four of the six experiments, the exceptions being 2002 late planted and 2003 early planted. This suggests that soybean aphid populations were not large enough to cause yield losses in these two experiments. The R3 spray treatment increased yield in three of the six experiments (2001 late planting, 2002 early planting, and 2003 late planting), the R2 spray treatment increased yield in two of six experiments (2001 and 2003 late plantings), and the V1 application increased yield over the control in the 2001 late-planted experiment. Results suggest that when aphid populations are high insecticide applications made at R2 and R3 plant stages are most effective in preventing yield loss.