This study aimed to determine the morphological differences of three-part proximal humerus fractures, the group in which plate screw fixation is most frequently used, and to evaluate the functional and radiological results of the methods applied for different subgroups. Twenty-nine patients (6 males and 23 females) with three-part proximal humerus fractures were in the study, with an average age of 64. The patients were in three groups according to their fracture types. Group 1 included eight patients with valgus impaction fracture. Group 2 included eleven patients with easily achieved stability after reduction. Group 3 consisted of ten patients with procurvatum varus angulation, a significant displacement between fragments, and in whom medial cortical continuity was not maintained without fixation. All patients underwent surgery with a minimally invasive deltoid split approach method and locked ana-tomic plate screw osteosynthesis. In group 1 patients, the space in the area where valgization is present in the head was filled with cortico-cancellous allografts. No grafting or metaphyseal compression took place in Group 2 patients. In group 3 patients, the metaphyseal compression technique was applied to the bone defect area. Cephalodiaphyseal angles (CDA) were measured at the postoperative and final follow-up. The constant Murley score made the functional evaluation. The patients were followed for an average of 27.6 months, and the union was present in all patients for an average of 3.6 months. Early screw migration was present in three patients, and late screw migration was in one patient. There were twenty-four excellent and 5 good results. CDA decreased from 139.42° to 136.13°. A statistically significant difference was present between the values of Groups 2 and 3 in the final control CDA of the groups. In this study, the functional scores of grafting stable valgus-impacted fractures and metaphyseal compression of unstable fractures with insufficient medial support were as good as stable 3-part fractures. Considering neer type 3 fractures should be evaluated with their subgroups, and fixation and stability-enhancing solutions specific to the groups are essential.