Sepsis and septic shock, which are often caused by pneumonia, impact millions of people every year. Despite adequate antibiotic therapy, mortality remains high, up to 45% in septic shock, which is characterized by an inappropriate, excessive immune response of the host. Moreover, critical illness-related corticosteroid insufficiency often coexists. Against this background, several trials and meta-analyses evaluated corticosteroid therapy as adjuvant therapy with heterogeneous results. Indeed, before 2000, high-dosage, short courses of corticosteroid treatment resulted in no benefit on mortality and a higher rate of adverse events. After 2000, thanks to a deeper understanding of the pathophysiology, low-dosage with longer courses of treatment were tested. With this regimen, a faster decrease in inflammation and faster resolution of shock, with a low rate of mild adverse events, was demonstrated although no clear effect on mortality was shown. To date, guidelines on sepsis and septic shock and guidelines on severe community-acquired pneumonia suggest corticosteroid use in selected patients. Furthermore, by utilizing latent class analysis, phenotypes of sepsis patients who benefit the most from corticosteroid treatment were recently identified. Future research should be guided by a precision medicine approach to identify adequate dosage and duration of corticosteroid treatment for appropriate patients. This article is freely available.