Research on psychotherapy in general, and psychodynamic therapy in particular, has consistently found that the quality of the alliance predicts the outcome of treatment. Little empirical support exists for the notion that the alliance sets the stage for techniques to be most effective, although adequate studies of this interaction between techniques and alliance are scarce. Research suggests that that the alliance is influenced in part by preexisting patient characteristics and in part by the quality of therapist interventions. Limitations of the empirical literature on the alliance are presented and some suggestions for future research are given.