Activation of the complement system by tumor cells was long believed to only benefit the host. Overexpression of complement inhibitors by many tumor cell types and results obtained in several experimental animal models were all in agreement with this hypothesis. However, recent reports imply that the situation is more complex than initially believed and that under certain circumstances tumor cells may use complement to their own advantage, e.g. by recruitment of suppressor T cells or promoting local angiogenesis. Such a dual role of complement may also be apparent when considering the effect of therapeutic monoclonal antibodies (mAb) used to successfully treat B cell malignancies, such as CD20 mAbs. Some argue that besides direct tumor cell killing by mAbs, two main immune effector mechanisms, complement dependent cytotoxicity (CDC) and antibody dependent cellular cytotoxicity (ADCC), may be competing with each other. Experiments aiming at answering the question whether complement is our friend or foe in mAb therapy ended up with seemingly contradictory conclusions. Herein, we revisit the existing knowledge on this pivotal issue based on rituximab and other anti-CD20 mAb as a model of therapeutic agents.