The pretreatment quality of intimate relationships can promote or interfere with couple therapy for posttraumatic stress disorder (PTSD) treatment response. We tested whether baseline relationship satisfaction predicted clinical and process outcomes in two dyadic treatments for PTSD. Using data from a randomized trial comparing brief cognitive behavioral conjoint therapy (bCBCT) for PTSD to PTSD family education (PFE) among 137 military veterans and their partners (N = 274, Mage = 42.3 years, 46.7% White, 81.0% male veteran partner), we examined whether baseline relationship satisfaction (Couples Satisfaction Index; CSI-32) predicted change in PTSD symptom severity (Clinician Administered PTSD Scale for DSM-5; CAPS-5), psychosocial functioning (Brief Inventory of Psychosocial Functioning; B-IPF), and relationship satisfaction at posttreatment and 6-month follow-up. We also explored associations with process outcomes (working alliance, treatment satisfaction, dropout). In both treatment conditions, neither partner's baseline CSI-32 score moderated change in veteran CAPS-5 or B-IPF score or any process variable. However, baseline CSI-32 scores moderated both partners' CSI-32 score change during bCBCT and PFE; participants who scored in the distressed range at baseline (n = 123) experienced significant improvements in relationship satisfaction, β = .199, whereas there was no change among those in the nondistressed range at baseline (n = 151), β = .025. Results suggest bCBCT and PFE are effective in improving PTSD symptoms and psychosocial functioning regardless of whether a couple is experiencing clinically significant relationship distress; further, these treatments improve relationship satisfaction for the most distressed individuals.