The oral ¹³C-bicarbonate technique (o¹³CBT) can be used for short-term measurements of CO₂ production (RCO₂) and energy expenditure (EEx). The method relies on appropriate estimates for the respiratory quotient (RQ) and recovery factor (RF) of ¹³C. Four Retriever dogs were included in four experiments to validate the o¹³CBT against indirect calorimetry (IC), and determine RQ and RF; Expt. 1: feeding different protein:fat:carbohydrate ratios [in % of metabolisable energy]: 25:33:42 in a maintenance (Mnt.) diet; 38:26:36 in a high-protein high-fibre (H_Fibre) diet and 27:56:17 in a high-fat (H_Fat) diet, simultaneously with start of measurements (T₀); Expt. 2: the Mnt. diet at T₀ or 4 h postprandial (T₄); Expt. 3: T₄ at different ambient temperatures, 22°C and 15°C; Expt. 4: T₄ after 1 h physical activity. The RCO₂ and EEx were determined from the respiration chamber measurements made simultaneously with IC and the o¹³CBT (o¹³CBT_online), and in Expts. 1 and 2, also on two consecutive days using o¹³CBT with collection of breath into breath bags (o¹³CBT_breathbags). The RQ values obtained at T₀ reflected dietary compositions, with the highest least square mean (LSM) of 0.954  for the Mnt. diet, 0.905 for the H_Fibre and 0.877 for the H_Fat diet (p < 0.05). An increased interval between meal and measurement period decreased RQ significantly (p < 0.05) in Expt. 2, LSM being 0.954 at T₀ and 0.909 at T₄. Ambient temperature (Expt. 3) and physical activity (Expt. 4) did not influence postprandial RQ. The RF values were not significantly affected by diet (Expt. 1). Measurements starting at T₀ (Expt. 2) resulted in higher (p < 0.05) RF values than at T₄ (LSM = 0.971 and 0.836, respectively). The ambient temperatures (Expt. 3) did not influence postprandial RF. However, when dogs were physically active prior to measurements (Expt. 4), RF values (LSM = 1.019) were higher (p < 0.05) than when resting only (LSM = 0.836). Calculations based on RQ and RF determined in each experiment resulted in RCO₂ and EEx values which were not different regardless of method used, except for Expt. 1 where EEx-values [kJ · kg BW^–0.75 · d^–1] were higher (p < 0.05) when measured with o¹³CBT_breathbags (460) than by IC (421) and o¹³CBT_online (420). Provided accurate RQ and RF values, the o¹³CBT_breathbags can be used as an independent and minimally invasive research tool to determine EEx in dogs under carefully standardised conditions.