There is a well-established bidirectional, negative association between couple satisfaction and depressive symptoms. Yet, a family systems perspective emphasizes the role of the therapist in interrupting this recursive cycle between couple satisfaction and depressive symptoms. The current study utilized longitudinal data to explore the bidirectional associations between depressive symptoms and couple satisfaction, moderated by the therapeutic alliance over the course of therapy. The study included 108 couples participating in couple therapy at a university training clinic. Couples rated their depressive symptoms and couple satisfaction separately before the intake session and at the end of the fourth session, and they also reported their individual therapeutic alliance with the therapist at the end of the second and third sessions. Actor-partner interdependence moderation model analysis revealed several moderation effects. In general, with low therapeutic alliance, couples with higher initial symptoms (such as depressive symptoms and low couple satisfaction) reported more severe symptoms at the fourth session, compared to those who had fewer initial symptoms. The moderating effect of alliance on a couple's symptoms was found both among individuals, and between partners. Systemic clinical implications and suggestions for future research are discussed.