Forty-five Holstein dairy cows (147 d-in-milk and 544 kg BW on average) were assigned to a 3 × 3 Latin square design replicated 15 times to evaluate the effect of a sensory additive (ProEfficient, PE; Lucta SA, Barcelona, Spain) on milk response to concentrate supplementation on pasture. Cows were blocked by parity in 15 blocks and within blocks randomly assigned to one of 3 treatments: 0 kg/d concentrate (0C); 5 kg/d control concentrate (CC); and 5 kg/d CC with 30 g/d of PE (PEC). Cows grazed a kikuyu (Pennisetum clandestinum) pasture for 84 d in 3 28-d periods. Targeted pasture allowance was 35 kg DM/cow. Pasture averaged 24.2% CP, 55.5% NDF, and 60.3% in vitro DM digestibility (IVDMD). Concentrates were fed twice daily at milking. Concentrate (44.3% corn, 11.2% wheat bran, 10.0% corn gluten feed, 8.6% rice meal, 8.4% soybean meal, 8.0% of sunflower meal, 9.5% mineral premix) averaged 18.4% CP, 17.8% NDF, and 85.5% IVDMD. Data were analyzed using a mixed model that included the fixed effects of treatment, block, period, their 2-way interactions, and the random effect of cow within block. On average, supplementation with the starch-based concentrate increased (P < 0.05) milk yield at 6.2 kg/d compared with 0C (15.9 vs. 22.1 kg/d, SEM 0.51). However, cows supplemented with PEC produced more (P < 0.05) milk than CC cows (22.7 vs. 21.4 kg/d, respectively; SEM 0.51). Milk production response to concentrate supplementation was improved at 15% (P < 0.05) by the addition of PE into the concentrate (1.32 kg vs. 1.15 kg milk/kg concentrate for PEC and CC, respectively). Concentrate supplementation reduced (P < 0.05) milk fat percentage (3.69 vs. 3.93%, SEM 0.065) and milk urea N (17.0 vs. 19.3 mg/dl, SEM 0.45). Milk fat and protein yield were increased (P < 0.05) by concentrate supplementation (0.794 vs. 0.620 kg/d, SEM 0.022 and 0.686 vs. 0.497 kg/d, SEM 0.016; respectively) compared with CC. The milk response triggered by PE might be associated with a reduction in the substitution rate of CC for pasture.