This paper investigates the effect of screw reinforcement on the capacity of timber members under compression perpendicular to the grain. The predictions of existing capacity models are compared to the experimental results of 39 timber specimens, distinguished by different load, screw and geometric configurations. Current capacity models assume two failure mechanisms, mainly characterized by their location, i.e. the contact area of the applied load (first mode) or the screw tips (second mode). However, the experimental tests reveal that the second mode never occurs despite the model predicting the occurrence of the second mode in more than half of the tested specimens. Additionally, the experimental tests show the fallacy of existing models in accurately estimating the capacity associated with the second failure mode. Parallely, the model appears to be relatively conservative for the first failure mode.