Tranexamic acid (TXA) in traumatic haemorrhage reduces death from bleeding. However, its effect on the coagulation system after trauma is unclear. We aimed to assess the effect of TXA on functional clotting dynamics in patients with suspected traumatic haemorrhage. Adult trauma patients for whom the major haemorrhage protocol was activated within a single major trauma centre were prospectively recruited. Blood was drawn on admission to the emergency department and during the acute bleeding phase (after transfusion of 4, 8, and 12 red blood cell [RBC] units) for thromboelastometry. Patients who received TXA before admission blood draw (TXA group) were compared with patients who did not (no TXA group). Proteins of coagulation and fibrinolysis were measured in a subset of patients. 222 patients were included (109 TXA group, 113 no TXA group) for functional clotting analysis, and 162 patients were included (60 TXA, 102 no TXA) for protein analysis. The proportion of patients with trauma-induced coagulopathy was lower in the TXA group than in the no TXA group throughout the acute bleeding phase (after 12 RBC units 67% [6/9] vs 100% [9/9] with coagulopathy, p=0·21). TXA abolished the functional hyperfibrinolysis of trauma-induced coagulopathy on admission (0% [0/109] vs 20% [23/113], p<0·0001) and during bleeding (after 12 RBC units 0% [0/9] vs 22% [2/9], p=0·47). Despite massive blood transfusion, clot strength was protected in the TXA group compared with the no TXA group (after 12 RBC units, median maximum clot firmness 53 mm [IQR 48–56] vs 44 [29–48], p=0·008). TXA switched off functional lysis even in the presence of extremely high concentrations of plasmin-antiplasmin complex. Patients treated with TXA had lower D-dimer levels after massive haemorrhage (after 12 RBC units 12·1 μg/mL [5·8–16·8] vs 30·4 [22·7–55·0], p=0·005). 24 h mortality was lower in the TXA group than in the no TXA group (7% [8/109] vs 23% [21/90], p=0·002) but not 28 day mortality. Our study shows that early TXA is associated with augmented clot strength, avoidance of hyperfibrinolysis, reduced trauma-induced coagulopathy, and lower 24 h mortality in patients with traumatic haemorrhage. None.