A multi-site pilot randomized controlled trial of Congruence Couple Therapy (CCT) for problem gambling was conducted in Ontario and Alberta, Canada from 2009 to 2011. The purpose was to assess the feasibility of a full trial and to identify methodological modifications to enhance future trials. The sample (N = 30; 15 couples) consisted of 66% male gamblers and 34% female. Mean age of sample was 49.1 years. Baseline mean DSM-IV gambling score was 8.7/10. Retention of the treatment couples was 89% at 2-month follow-up. Retention of control couples was 78%. A randomized controlled design compared the status of couples in treatment condition to control condition. Treatment couples received 12-week CCT while control couples received three brief check-ins over 12 weeks. No significant difference was found between treatment and control group at baseline on all measures. At (1) week 12 post-treatment, and (2) week 20 follow-up, significant treatment effects were found for gambling symptoms (p = 0.008; p = 0.041), mental distress (p = 0.001; p = 0.035), and family systems function (p = 0.023; p = 0.054) between treatment and control group. Within group changes for treatment couples over time were significant for mental distress (p = 0.000), dyadic adjustment (p = 0.002), and family systems function (p = 0.000). On similar measures, control group showed non-significant improvement. Future methodological changes, advantages and disadvantages of multi-site partnerships with community treatment agencies are discussed. Of interest is that control participants showed unintended improvement. CCT as a treatment was favourably accepted by counselors, problem gamblers and their spouses. Positive outcome trends ranging from small to large effect size on key measures indicate that a full-scaled trial will require approximately 140 couples and is an investment worth pursuing.