Several treatment-related challenges exist for gambling disorder, in particular at-scale dissemination in health care settings. This study describes the introduction of a newly developed internet-delivered cognitive behavioral therapy (iCBT) program for gambling disorder (GD), provided with therapist support in routine addiction care, in a nationally recruited sample in Sweden. The study details the introduction of the iCBT program, evaluates its effectiveness and acceptability, and compares registry outcomes among iCBT patients with other patients with GD at the clinic who received face-to-face psychological treatment as usual. The study site was the Stockholm Addiction eClinic, which offers digital interventions for addictive disorders in routine care. The iCBT program was introduced nationally for treatment-seeking patients through the Swedish eHealth platform. After approximately 2 years of routine treatment provision, we conducted a registry study, including ordinary patients in routine digital care (n=218), and a reference sample receiving face-to-face psychological treatment for GD (n=216). A statistically significant reduction in the Gambling Symptom Assessment Scale scores during the treatment was observed (B=-1.33, SE=0.17, P<.001), corresponding to a large within-group Cohen d effect size of d=1.39. The iCBT program was rated high for satisfaction. A registry-based survival analysis, controlling for psychiatric comorbidity, showed that patients receiving iCBT exhibited posttreatment outcomes (re-engagement in outpatient addiction care, receiving new psychiatric prescriptions, enrollment in psychiatric inpatient care, and care events indicative of contact with social services) similar to comparable patients who underwent face-to-face treatment-as-usual. A lack of randomized allocation notwithstanding, the iCBT program for GD evaluated in this study was well-received by patients in routine addiction care, was associated with the expected symptom decrease during treatment, and appears to result in posttreatment registry outcomes similar to face-to-face treatment. Future studies on treatment mechanisms and moderators are warranted. RR2-10.1186/s40814-020-00647-5.