The effects of supplementing grass with a high-energy, low-protein concentrate on intake, rumen fermentation, rumen microbial communities, apparent total tract digestibility and N-balance in beef cattle, and rumen fermentation and methane output in vitro, were studied. Sixteen Charolais steers (mean live weight (s.d.) 564 (24.3) kg)), were used in a completely randomised block design experiment, and offered either zero-grazed perennial ryegrass (Lolium perenne L.) ad libitum (GO), or the same grass with 2.25 kg citrus pulp dry matter (DM) daily (GCP). During the measurement phase, DM intake was restricted at 0.85 of their pre-determined intake. The same diets were incubated in an eight-vessel RUSITEC system. Supplementation decreased grass DM intake (6.59 vs. 7.54 kg/d) but increased total DM intake by 0.95 kg/d (P < 0.01). Rumen pH (P = 0.05) and NH3 concentration (P < 0.05) decreased, and total volatile fatty acid (VFA) concentrations (P < 0.01), molar proportions of propionate (P = 0.08) and butyrate (P < 0.05) increased for GCP compared to GO. Supplementation significantly altered the overall rumen microbial community structure compared to GO, driven principally by changes in relative abundance of Prevotella, Methoanobrevibacter and Lachnospiraceae genera. Plasma urea concentration was lower (P < 0.05) for GCP than GO. There was no effect (P > 0.10) of supplementation on apparent total tract digestibility of DM, OM and ADFom, but aNDFom digestibility tended to be lower (P = 0.07) and crude protein (N × 6.25) digestibility was lower (P < 0.01) for GCP compared to GO. Nitrogen intake (200 g/d) did not differ (P > 0.05) between treatments. Faeces N loss was 9 g/d greater (P < 0.01), and urine (P < 0.01) and total (P < 0.05) N losses were 22 and 13 g/d lower, respectively, for GCP compared to GO. In terms of N-balance (g/kg N intake), faecal N excretion was greater and urinary N excretion was lower (P < 0.01) resulting in a greater N use efficiency (P = 0.06) for GCP than GO. In vitro rumen NH3 concentrations were lower (P < 0.01) for GCP compared to GO, whereas in vitro rumen pH, lactic acid, VFA concentrations and molar proportions, and methane and total gas output did not differ (P > 0.05) between treatments. Supplementing grass with a low-protein, high-energy feed may be a strategy to reduce urinary and total N excretion in beef cattle grazing temperate grass pasture.