The aim of this retrospective observational study was to evaluate whether adding liraglutide to lifestyle changes, metformin (Met) and testosterone replacement therapy (TRT), by means of improving weight and glycaemic control, could boost erectile function in type 2 diabetic obese men with overt hypogonadism and erectile dysfunction (ED) in a 'real-life setting'. Forty-three obese, diabetic and hypogonadal men (aged 45-59 years) were evaluated because of complaining about the recent onset of ED. They were subdivided into two groups according to whether hypogonadism occurred after puberty (G1; n = 30: 25 with dysfunctional hypogonadism and 5 with acquired hypogonadotropic hypogonadism) or before puberty (G2; n = 13: 10 with Klinefelter's syndrome and 3 with idiopathic hypogonadotropic hypogonadism). Both G1 and G2 patients were given a combination of testosterone (T) [testosterone undecanoate (TU) 1000 mg/every 12 weeks] and Met (2000-3000 mg/day) for 1 year. In the poor responders (N) to this therapy in terms of glycaemic target (G1N: n = 16; G2N: n = 10), liraglutide (L) (1.2 μg/day) was added for a second year, while the good responders (Y) to T + Met (G1Y: 14/30 and G2Y: 3/13) continued this two drugs regimen therapy for another year. All patients were asked to fill in the International Index of Erectile Function (IIEF 15) questionnaire before starting TU plus Met (T1) and after 12 months (T2) and 24 months (T3) of treatment. Patients underwent a clinical examination and a determination of serum sex hormone binding globulin (SHBG), total testosterone (T) and glycosylated haemoglobin (HbA1c) at T1, T2 and T3. At T2, each patient obtained an improvement of ED (p < 0.01) and of the metabolic parameters without reaching, however, the glycaemic goals [HbA1c = >7.5% (>58 mmol/mol)], while T turned out to be within the range of young men. L added to TU and Met regimen in G1N and G2N allowed these patients to reach not only the glycaemic target [HbA1c = <7.5% (<58 nmol/mol)] and a significant reduction in body weight (p < 0.01), but also a further increase in SHBG (p < 0.05) and T (p < 0.01) plasma levels as well as a significant increment of IIEF score (T3). Conversely, at T3 G1Y and G2Y, who received the combined therapy with TRT and Met for the second year, showed a partial failure of that treatment given that there was no improvement of the IIEF score and they showed a significant rise in serum HbA1c (p < 0.05) and weight (p < 0.04) compared with the assessments at T2. These results suggest that TRT could improve clinical and metabolic parameters in obese, type 2 diabetic men with ED and overt hypogonadism (independently of when T deficit occurred). Furthermore, in case of insufficient metabolic control the addition of L to TRT and Met regimen allows to achieve serum T levels in the range of healthy men, as well as to reach glycaemic target and to lower weight, leading to a considerable improvement of ED.