This letter details and discusses the next-to-leading order QCD corrections to t-channel electro-weak W + bj production, where finite top-width effects are consistently taken into account. The computation is done within the aMC@NLO framework and includes both resonant and non-resonant contributions as well as interferences between the two. Results are presented for the LHC and compared to those of the narrow-width approximation and effective theory approaches.It is of no surprise therefore that over the last three decades much effort has been invested in providing accurate predictions for single-top cross sections. Stable single-top production was first discussed at the leading order (LO) in [4]. Next-to-leading order (NLO) QCD corrections in the five-flavour (5F) scheme were first computed in [5] and in the four-flavour (4F) scheme in [6,7], whilst electroweak (EW) corrections were investigated in [8]. Furthermore, soft and collinear gluon resummation has been studied in [9,10,11,12] and stable single-top at NLO matched to parton showers has been achieved in [13,14,15] for both the 5F and 4F schemes. Incorporating the decay of the top quark, t → W + b, as part of the hard process matrix elements was accomplished via the narrow-width approximation (NWA) in [16,17,18], in which NLO corrections to both production and decay were included. Going beyond the NWA, a study of off-shell and non-resonant effects at LO was performed in [19] and final-state non-factorizable corrections in the s-channel were examined in [20]. In [21,22], effective theory (ET) techniques were employed to relax the assumption of an on-shell top quark in the amplitudes, allowing for a systematic study of finite top-width (Γ t ) effects in the resonant regions of phase space.The full computation of α s -corrections to EW W + bj production (i.e. the process that includes single-top production) has so far been missing in the literature. Given the opportunity for precision top physics provided by the LHC, it is of phenomenological interest to understand the effects of off-shell top quarks versus top quarks in the NWA. In this letter we present such a computation for the tchannel process in the 5F scheme, and make a direct comparison to results in the NWA as well as to those obtained by using the ET method. The s-channel and W t-production processes are not considered here.The expectation is that for inclusive observables off-shell effects are small [23,24,25] (i.e., parametrically of O(Γ t /m t )), whilst they should be noticeable in the case of less inclusive observables, such as the invariant or transverse masses of the reconstructed top. Indeed, differences between NWA and ET or off-shell calculations have already been highlighted for 5F scheme single-top [21,22] and top-pair [26,27,28,29] production. We shall confirm these findings here.This paper is organized as follows. In Sect. 2 we briefly discuss the NWA, ET, and off-shell approaches in view of their application to single-top production; results for total rates and for a few selected distri...