The objective of this experiment was to assess a new method to determine the bioavailability (BA) of rumen-protected (RP)His, RPLys and RPMet products by determining rumen escape (RE) and fecal excretion of undigested AA from RPAA. Eight lactating Holstein cows (79 ± 21 days in milk, 53.0 ± 7.0 kg/d milk yield), four of which rumen-cannulated, were used in a replicated 4 × 4 Latin square design experiment with four, 26-day experimental periods and a preceding 22-day background (BG) period. Four combinations of nine commercial and experimental RPAA (HisA, HisB, LysA, LysB, LysC, MetA, MetB, MetC, MetD) were fed daily to supply 20, 25, and 35 g/day of digestible (d)His, dLys and dMet, respectively. The following treatment combinations were used: (1) HisALysAMetA, (2) HisBLysBMetB, (3) LysCMetC and (4) LysCMetD. Spot sampling of feces was performed during the BG period to establish basal levels of AA in feces. Total fecal collection and blood sampling were performed during the last three days of each experimental period. Rumen escape fraction of each RPAA product was determined in situ and was greater for HisA (0.90) than for HisB (0.64), ranged from 0.33 to 0.85 (SEM = 0.017) for RPLys and from 0.53 to 0.95 (SEM = 0.017) for RPMet. Apparent post-ruminal digestibility of AA from RPAA was calculated as [(RE of AA, g/day – BG corrected fecal AA output, g/day) ÷ RE of AA, g/day]. Digestibility was similar between the two RPHis products (0.85 and 0.90). Apparent post-ruminal digestibility of Lys from RPLys products varied from 0.30 to 0.79 and digestibility for RPMet varied from 0.85 to 0.96. Bioavailability was calculated as: (RE, g/g × AA digestibility, g/g) × 100 and was greater for HisA compared with HisB (76.1 and 57.9 % respectively), varied from 9.63 (LysC) to 67.0 % (LysA) for the RPLys and was lowest for MetC (49.3 %) and greatest for MetD (91.9 %) among the RPMet products. Plasma His concentration was greater for HisA than for HisB and plasma Met concentration was greater for MetD compared with the other three RPMet products, reflecting estimated BA of the RPAA products. In contrast, plasma Lys concentration did not differ among RPLys products. This study showed that a method using fecal AA output in combination with RE of RPAA can reveal differences in BA between RPAA products. This is a first step in establishing the fecal AA excretion technique for determination of BA of RPAA and warrants further validation.