The purpose of this study was to assess the pharmacokinetics of robenacoxib (RX), a COX-2 selective non-steroidal anti-inflammatory drug, in goats after single intravenous (IV), subcutaneous (SC) and oral (PO) administrations. 5-month-old healthy female goats (n = 8) were used. The animals were subjected to a three-phase, two-dose (2 mg/kg IV, 4 mg/kg SC, PO) unblinded, parallel study design, with a four-month washout period between the IV and SC treatment, and a one-week period between the SC and PO treatment. Blood was drawn from the jugular vein in heparinized vacutainer tubes at 0, 0.085 (for IV only), 0.25, 0.5, 0.75, 1, 1.5, 2, 4, 6, 8, 10 and 24 h. Plasma RX concentrations were measured using HPLC coupled to a UV multiple wavelength detector, and the data were pharmacokinetically analysed using ThothPro™ 4.3 software in a non-compartmental approach. Following IV administration, terminal elimination half-life, volume of distribution and total clearance were 0.32 h, 0.24 L/kg and 0.52 L/h/kg, respectively. For SC and PO, the mean peak plasma concentrations were 2.34 and 3.34 μg/mL at 1.50 and 0.50 h, respectively. The t1/2λz was significantly different between the IV and the extravascular (EV) administrations (0.32 h IV vs 1.37 h SC and 1.63 h PO), suggesting the occurrence of a flip-flop phenomenon. The significant difference in Vd values between IV (0.24 L/kg) and EV (0.95 L/kg SC and 1.71 L/kg; corrected for F %) routes might have also triggered the t1/2λz difference. The absolute average SC and PO bioavailability were high (98% and 91%, respectively). In conclusion, the IV administration of RX might not be suitable for goats, due to its short t1/2λz. The EV routes, however, appear to be convenient for the drug's occasional use.