Two methods are commonly used to describe the feeding behavior of wild primates, one based on the proportion of time animals spent feeding on specific plant parts ("time-based" estimates) and one based on estimates of the actual amounts of different plant materials ingested ('"weight-based" estimates). However, studies based on feeding time may not be accurate for making quantitative assessments of animals' nutrient and energy intake. We analyzed the diet of two groups of Alouatta pigra living in forest fragments using two different methods (time- and dry weight-based estimates), to explore how these alternative approaches impact estimates of (a) the contribution of each food type to the diet and (b) the macronutrient composition of the diet, including available protein (AP), non-protein energy (NPE), and total energy (TE) intake. We conducted behavioral observations (N = 658 hr and N = 46 full day focal follows), from August 2012 to March 2013. For each feeding bout, we estimated both time spent feeding and actual fresh- and dry-weight consumption by counting the number of food items ingested during the bout. Using time-based estimates, A. pigra showed a predominantly leaf-based diet. In contrast, weight-based estimates described combined a fruit and leaf-based diet. There were no differences between methods when estimating AP intake; however, we found significant differences while estimating NPE and TE intake. Time-based estimates provide us with important information such as the foraging effort spent on food items, trees, or patches, while weight-based estimates may provide more accurate information concerning nutrient and energy intake. We suggest that quantitative estimates of nutrient intake in a primate's diet be based on observations of wet and/or dry weight actually ingested rather than extrapolated from time spent feeding.