A growing body of research has shown that strategic feeding and nutritional management of preweaning heifers can have life-long effects on their performance. Our objective was to assess the performance of first-lactation heifers that were fed either accelerated milk replacer (MR; 3.98% CP, 2.56% fat) or non-saleable milk (3.59 ± 0.28% true protein; 4.12 ± 0.37% fat) that was either pasteurized (PM) or raw (RM). Heifers were randomly assigned to feeding treatments after birth and were fed 3 times daily either 1.4 L (calves <36.3 kg, until the target weight was achieved) or 1.9 L (calves ≥36.3 kg). After weaning, all heifers were managed uniformly. Monthly test day data were used to generate DHIA estimates of 305-d mature equivalent milk (MEM), fat (MEF), and protein (MEP) yields; predicted transmitted ability (PTA) estimates for these traits were also collected. Data were analyzed using mixed, regression, and logistic procedures of SAS and significance declared at P ≤ 0.05. The model included the main effect of milk treatments, amount fed daily at enrollment, and their interaction. The corresponding PTA was used as covariate for milk yield variables. The sample size (n = 152) provided sufficient power (80%) to detect statistical differences of 950 kg MEM between treatments before accounting for the covariate. Treatments did not influence (P > 0.10) the number of inseminations at first calving, age at first calving, days open, or retention in the herd by 36 mo of age (left/enrolled: 18/50, 18/50, 16/52 for MR, PM, and RM, respectively). Heifers fed 5.7 L/d produced more MEM (P = 0.04; + 695 kg), tended (P = 0.07) to produce more MEF (+26 kg), and produced more MEP (P = 0.05; +18 kg) compared with heifers fed 4.2 L/d, which may be related to either prenatal factors causing low BW or to the level of nutrition early in life. An interaction of treatment and feeding rate was observed for MEM and MEF (P = 0.07); within the RM-treatment, low-BW heifers produced less MEM and MEF compared with high-BW heifers (−1,705 kg MEM and −68 kg MEF, both P < 0.01). Although no overall treatment effects on first-lactation performance were observed, feeding RM may impair first lactation performance of low-BW heifers.