Purpose Glenohumeral osteoarthritis (OA) represents a challenging problem in young, physically active patients. It was the purpose of this investigation to evaluate the results of a pilot study involving glenoid resurfacing with a glenoid allograft combined with a hemiarthroplasty on the humeral side. Methods Between April 2011 to November 2013, 5 patients (3 men, 2 women, mean age 46.4, range 35-57) with advanced OA of the glenohumeral joint, were treated with a humeral head replacement combined with replacement of the glenoid surface with an osteochondral, glenoid allograft. Results Overall, clinically, there was one excellent, one satisfactory and three poor results. Mean preoperative subjective shoulder value (SSV) was 34% (range: 20-50%) and preoperative relative Constant-Murley-Score (CSr) was 43 points (range: 29-64 points). Three patients with poor results had to be revised within the first three years. Their mean pre-revision SSV and CSr were 38% (range: 15-80%) and 36 points (range: 7-59 points) respectively. One patient was revised 9 years after the primary procedure with advanced glenoid erosion and pain and one patient has an ongoing satisfactory outcome without revision. Their SSVs were 60% and 83%, their CSr were 65 points and 91 points, 9 and 10 years after the primary procedure, respectively. Mean follow-up was 7 years (2-10 years) and mean time to revision was 4 years (range: 1-9 years). Conclusion The in-vivo pilot study of a previously established in-vitro technique of osteochondral glenoid allograft combined with humeral HA led to three early failures and only one really satisfactory clinical outcome which, however, was associated with advanced glenoid erosion. Osteochondral allograft glenoid resurfacing was associated with an unacceptable early failure rate and no results superior to those widely documented for HA or TSA, so that the procedure has been abandoned. Level of evidence Level IV, Case Series, Treatment Study.