The optimization of process parameters of the nanofluid preparation process for maximum stability and high heat transfer is an active and important area of research. In this work, the effect of the surfactant material, surfactant weight, and ultrasonication time are studied on distilled water-based CuO, Fe3O4, and CuO+Fe3O4 nanofluids. Taguchi L9 orthogonal array was used for the design of the experiment and 9 samples were prepared using this array. The effect of each level of process parameter on the thermal conductivity is analyzed by calculating Signal to Noise Ratio (SNR) and optimum levels of these parameters are identified. The crucial role of stability in delivering high thermal conductivity nanofluids as predicted by SNR analysis is further confirmed using Analysis of Variance (ANOVA).