["Australian and New Zealand Journal of Family Therapy, Volume 47, Issue 3, September 2026. ", "\nABSTRACT\nWalk‐in single‐session models have the potential to provide timely access to family mental health support, while addressing service system resource constraints. Single‐session therapy models have demonstrated effectiveness across various contexts internationally, yet few studies have examined their impact on family and relational outcomes. This study aimed to compare the effectiveness of two family therapy interventions within the same service setting for families facing complex challenges, such as mental illness, substance use and trauma. One offered immediate access to a single online session; the other provided multiple sessions of therapy following a full intake assessment and a variable wait time. Sixty family members (30 per intervention) participated in the 12‐month follow‐up. Using a mixed‐methods design, measures of confidence, satisfaction and relational outcomes were collected, and in‐depth qualitative interviews conducted. At the 1‐year follow‐up for both interventions, sustained improvement in relational and family functioning outcomes was noted and did not differ significantly between groups. Similar confidence levels were reported about the continued effective management of the original concerns. Families in the single‐session group universally valued rapid access to whole‐of‐family care, and this appears to be a central contributor to positive outcomes. Although most family members reported positive outcomes from just one session, the benefit of longer‐term support pathways for some families was also evident. The findings of this relatively small study support the potential of single‐session, rapid‐access interventions to provide effective support for distressed families dealing with complex challenges. Timeliness and a well‐defined single‐session focus may create conditions of support that enable change at levels similar to traditional multi‐session family therapy. Further studies with larger sample sizes across different service contexts appear warranted.\n"]