The dialysis‐based definition of Delayed Graft Function (dDGF) is not necessarily objective as it depends on the individual physician’s decision. The functional definition of DGF (fDGF, the failure of serum creatinine to decrease by at least 10% daily on 3 consecutive days during the first week post‐transplant), may be more sensitive to reflect recovery after the ischemia‐reperfusion injury. We retrospectively analyzed both definitions in 253 deceased donor kidney transplant recipients for predicting death‐censored graft failure as primary outcome, using eGFR < 25 ml/min/1.73 m2 as a surrogate end‐point for graft failure. Secondary outcome was a composite outcome that included graft failure as above and also patient’s death. Median follow‐up was 3.22 2.38–4.21 years. Seventy‐nine patients developed dDGF (31.2%) and 127 developed fDGF (50.2%). Sixty‐three patients fulfilled criteria for both definitions (24.9%). At multivariable analysis, the two definitions were significantly associated with the primary HR (95%CI) 2.07 (1.09–3.94), P = 0.026 for fDGF and HR (95%CI) 2.41 (1.33–4.37), P = 0.004 for dDGF and the secondary composite outcome HR (95%CI) 1.58 (1.01–2.51), P = 0.047 for fDGF and HR (95%CI) 1.67 (1.05–2.66), P = 0.028 for dDGF. Patients who met criteria for both definitions had the worst prognosis, with a three‐year estimates (95%CI) of survival from the primary and secondary outcomes of 2.31 (2.02–2.59) and 2.20 (1.91–2.49) years for fDGF+/dDGF+, in comparison with the other groups (P < 0.01 for trend). fDGF provides supplementary information about graft outcomes on top of the dDGF definition in a modern series of kidney transplantation.