Insulin resistance and androgen excess, alongside anovulatory infertility, are the cardinal clinical and biochemical features of polycystic ovary syndrome (PCOS). Circulating androgen burden and metabolic dysfunction in PCOS are closely correlated, but an independent contribution of androgens per se to metabolic and other complications of PCOS remains poorly characterised. My work since 2012 has focused on delineating the distinct impact of androgens on metabolic function, with a particular focus on adipose tissue and insulin resistance. Adipose tissue is capable of androgen activation, and has a complex network of activating and inactivating enzymes. One of these enzymes, aldoketoreductase type 1 C3 (AKR1C3), activates the androgen precursor androstenedione to more potent testosterone. AKR1C3 expression is upregulated in subcutaneous adipose tissue from women with PCOS compared to BMI-matched controls. Using adipose tissue microdialysis, we have shown that PCOS women have significantly increased adipose concentrations of the active androgens testosterone and dihydrotestosterone compared to controls. Furthermore, using in vivo and in vitro studies, we have demonstrated direct effects of intra-adipose androgens on adipocyte lipid biology, with increased de novo lipogenesis and suppression of lipolysis promoting adipocyte hypertrophy. In other aspects of the presentation, I will discuss the relative contribution of the 11-oxygenated androgen synthesis pathway to circulating androgen burden and metabolic dysfunction in PCOS, which traditionally has been understudied in PCOS and other disorders of androgen excess. Data from a number of population-based studies will also be presented, which we have used to delineate the independent effects of androgen excess on metabolic disorders such as diabetes and non-alcoholic fatty liver disease, as well as on less well characterised potential complications of androgen excess such as obstructive sleep apnoea and idiopathic intracranial hypertension.