Acquired coagulation defects are characterized by a decrease of both pro- and anti-coagulants. Because of this, we hypothesise that global tests, such as the prothrombin and partial thromboplastin times (PT and APTT), might be unsuitable for their investigation. Indeed, these tests are not good predictors of bleeding in acquired coagulopathies as they are in the congenital ones. This article discusses the possible reasons for this, using cirrhosis and the neonatal period as epitomes of acquired coagulation defects. Both display normal thrombin generation in the presence of thrombomodulin, in spite of prolonged PT and APTT. We surmise that, because of their design, the PT and APTT are responsive to thrombin generated as a function of pro-coagulants, but much less to thrombin inhibited by the anti-coagulants, especially protein C, which is activated to a limited extent in the absence of thrombomodulin. In conclusion, the PT and APTT can tell us whether or not a patient is deficient in one or more pro-coagulants, but not whether this deficiency is counterbalanced by a parallel deficiency of anti-coagulants. Thrombin generation assays are more suitable than PT and APTT for use in acquired coagulation defects.