Our goal was to determine the importance of including body weight (BW) loss in assessing the response of cows to low protein diets. 169 Holstein cows in mid-lactation (92 primiparous) and 69 in late lactation (42 primiparous) were fed diets high or low in protein in crossover designs with 2 28-d treatment periods. Mid-lactation diets were 18% or 14% CP, and late-lactation diets were 16% or 12% CP. All diets had adequate RDP with expeller soybean meal added for high protein diets. Cows were milked 2×/d; intake and milk yield were recorded daily. Milk composition was measured 2 d/wk, and BW was measured 3d/wk. Historical prices of corn and soybean meal, milk class and component prices, and cull cow values were used to set financial parameters. Captured energy, captured protein, and gross income from milk production and BW change were calculated for each cow in each diet. Protein responses within each lactation stage were analyzed with the MIXED Procedure (SAS 9.4), including fixed effects of experiment, diet, parity, and period nested within experiment, and random effect of cow nested within experiment and parity. In mid lactation, reducing protein from 18 to 14% saved $0.73 per cow in daily feed cost but resulted in daily losses of: 1) 2.9 Mcal milk energy and 1.8 Mcal body energy, 2) 0.14 kg milk protein and 0.03 kg body protein, 3) $1.40 milk income and $0.29 body salvage value (all P < 0.05), so that 38% of the total energy loss, 18% of total protein loss, 17% of gross income loss, and 24% of income over feed cost (IOFC) loss were due to BW loss. In late lactation, reducing protein from 16 to 12% saved $0.60 in daily feed cost but resulted in daily losses of: 1) 2.3 Mcal milk energy and 2.2 Mcal body energy, 2) 0.11 kg milk protein and 0.05 kg body protein, 3) $1.07 milk income and $0.38 body salvage value (all P < 0.05), so that 50% of total energy loss, 31% of total protein loss, 26% of gross income loss, and 45% of IOFC loss were due to BW loss. In addition, the importance of BW loss was greater for primiparous than multiparous cows. In conclusion, BW loss was an important part of the response to low protein, and we suggest that it should not be neglected when assessing responses to dietary protein.