Background The majority of potentially preventable deaths in trauma are due to uncontrolled hemorrhage and occur early after injury. How major bleeding is defined is integral to early identification and treatment of this group of high‐risk patients. However, there is no consensus on a definition of major bleeding in trauma. The aim of this Delphi study was to develop a consensus definition for research, with transfusion used as a surrogate marker of bleeding. Study Design and Methods Trauma experts from three international groups were invited to take part in an online Delphi survey. Over the course of four rounds, the group developed a number of definitions of major bleeding and reached consensus on a new definition. Results Forty‐four trauma experts agreed to become members of the Delphi panel, and 30 of 44 (68%) completed all four rounds. The panel agreed to exclude the historical massive transfusion definition (≥10 units of red blood cells within 24 hours). Consensus was reached on a new definition for use in clinical research: 4 or more units of any blood component within 2 hours of injury. Conclusion This Delphi process has yielded a pragmatic transfusion‐based definition of major bleeding. The consensus definition differs from historical definitions: a shorter time frame to reflect the acuity of bleeding, and multiple blood components in keeping with a balanced approach to resuscitation. The definition developed may be best suited to mature trauma systems (reflecting the demographics of the expert panel), and could be used to guide registry data recording and to characterize patients at risk of major bleeding.