In neonatal patients, bleeding is a multifactorial event in which several factors may play a pathogenic role. Among these, thrombocytopenia is often considered a risk factor for bleeding, although a causal relationship has never been demonstrated. In fact, major bleeding mainly occurs in non-thrombocytopenic newborns and thrombocytopenic newborns rarely experience major bleeding. Therefore, parameters other than platelet count might better assess the hemostatic function and define bleeding risk. Historically, neonatologists aimed to reduce the risk of bleeding by administering platelet transfusions. However, recent studies demonstrated that transfusing newborns at higher threshold is associated with an increased risk of death, bleeding, bronchopulmonary dysplasia and neurodevelopmental impairment. The mechanism behind this association is not known and various hypotheses have been proposed, including the non-hemostatic effects of adult-derived platelets transfused into neonates. Alternatively, the rapid volume expansion caused by a platelet transfusion might cause hemodynamic instability and cardiocirculatory overload. Guidelines about platelet transfusions should now include this recent evidence and adopt more stringent thresholds. Future research should focus on finding alternative or improved transfusion products more suitable for newborns.