Donkey slaughter in West Africa has received limited scientific attention, despite increasing over recent years. This study aimed to explore factors affecting donkey welfare, both ante-mortem and at slaughter, in the Upper East region of Ghana. A total of 134 donkeys at five different slaughter points were assessed using animal-based indicators. Slaughter involved either blunt force trauma (BFT) using a wooden pole (1 location) or a metal hammer (2 locations); or non-stun ventral neck incision (VNI) (2 locations). Time to loss of consciousness (or return) was assessed, with behavioural and brainstem signs of sensibility/consciousness. Negative human–animal interactions occurred ante-mortem, with animals struck multiple times and handled using aversive methods. Donkeys hit on the head with the wooden pole were five times more likely to be ineffectively concussed (p < 0.001, OR: 5.4, CI: 1.9–15.4) compared to the hammer and took significantly longer to lose consciousness (p < 0.001). The mean time to loss of corneal reflex for those hit by wooden pole was 166.9 s (SD 21.1; range: 79–425 s), compared to just one animal displaying corneal reflex for 59 s after being hit by metal hammer. For those animals slaughtered through VNI, corneal reflex was observed for a mean time of 96.5 s (SD 4.3, range 26–164 s). The findings of this study highlight major concerns regarding the methods used for donkey slaughter in the Upper East region in Ghana, particularly regarding aversive ante-mortem handling and prolonged time to loss of consciousness during VNI and ineffective BFT in inducing unconsciousness by concussion.