Introduction: Coronavirus disease (COVID-19) has recently dominated scientific literature. Incomplete understanding and a lack of data concerning the pathophysiology, epidemiology, and optimal treatment of the disease has resulted in conflicting recommendations. Adherence to existing guidelines and actual treatment strategies have thus far not been studied systematically. We hypothesized that capturing the variance in care would lead to the discovery of aspects that need further research and—in case of proven benefits of interventions not being performed—better communication to care providers. Methods: This article is based on a quantitative and qualitative cross-sectional mixed-methods online survey among intensive-care physicians in Germany during the COVID-19 pandemic by the CEOsys (COVID-19 Evidence Ecosystem) network, endorsed by the German Interdisciplinary Association for Intensive Care and Emergency Medicine (DIVI) conducted from December 3 to 31 December 2020. Results: We identified several areas of care with an especially high variance in treatment among hospitals in Germany. Crucially, 51.5% of the participating ICUs (n = 205) reported using intubation as a last resort for respiratory failure in COVID-19 patients, while 21.8% used intubation early after admission. Furthermore, 11.5% considered extracorporeal membrane oxygenation (ECMO) in awake patients. Finally, 72.3% of respondents used the ARDS-network-table to titrate positive end-expiratory-pressure (PEEP) levels, with 36.9% choosing the low-PEEP table and 41.8% the high-PEEP table. Conclusions: We found that significant differences exist between reported treatment strategies and that adherence to published guidelines is variable. We describe necessary steps for future research based on our results highlighting significant clinical variability in care.