Processed complementary foods (PCF) might mitigate several complementary feeding barriers in developing countries. Efficacy trials, however, have not shown substantial improvements in child growth, possibly due to inadequate formative research to assess acceptability and identify pitfalls. Milk powder might improve palatability of PCF but incurs a higher cost. We compared the acceptability of an instant soy-rice PCF without (SR) and with (SRM) milk powder. Best practices for formative evaluation of PCF are not established. We therefore compared findings from randomized trials of SR vs. SRM in 1-d sensory tests (n = 71 mother-infant dyads) vs. Trials of Improved Practices (TIPs), a 2-wk in-home mixed methods evaluation (n = 54 dyads). TIPs included interviews, disappearance rates, observations, and 24-h dietary recalls to assess acceptance, consumption of the 50 g/d ration, and impact on diet. Although mothers preferred SRM to SR in the sensory tests, children in the TIPs consumed >50 g/d of SR (87 +/- 9 g/d) and SRM (89 +/- 8 g/d) with no difference between the foods (P = 0.55). Despite some replacement of family food, energy (574 kJ/d; P < 0.001) and protein (19 g protein/d; P < 0.001) increased in both groups. Mothers' preferences for milk, more sugar in SR, and preparation with hot water were concerns raised in the sensory tests that proved insignificant in TIPs. However, TIPs uncovered new concerns of overconsumption and food safety. We found milk did not improve the acceptability of the soy-rice PCF and recommend TIPs as a useful tool for formative research of PCF interventions.