Although there are several steel-to-timber joint types, there is still a lack of studies that more precisely characterize the components in the interaction between steel and timber in compression. Thus, this work aims to characterize the steel T-stub compressing the glued laminated timber (GLT) parallel to the grain component, focusing on formulating analytical expressions for design resistance, stiffness coefficient and deformation capacity. In order to validate the analytical models and increase the reliability of the component method, an experimental campaign was conducted through compression tests on 90 specimens composed of steel T-stubs with five different flange thicknesses under full and partial compression. It was identified that although GLT has lower compressive strength and lower stiffness than steel, the main failure mode in specimens studied was T-stub flange yielding. The proposed formulations were consistent with the experimental results and it can be used to predict the structural behavior of this component. The application of the component analytical variable elastic approach to a semi-rigid connection showed an increase in the accuracy for predicting its moment resistance. •The main failure mode observed was the T-stub flange yielding, due to small stiffness of the flange and to the concentration of stresses in the central region of the timber.•The component is characterized by proposing equations for resistance, stiffness coefficient, and deformation capacity. The precision of these parameters has been improved compared to experimental results and to equations proposed by other authors for joints with similar geometry.•Experimental evaluation of the steel T-stub compressing timber parallel to the grain component carried out with five different T-stub thicknesses and timber subjected to full and partial compression.