Background Based on the current state of research regarding the treatment in pediatric obsessive–compulsive disorder (OCD), cognitive behavioral therapy (CBT) (in severe cases with additional pharmacotherapy) is considered as the first-line treatment according to internationally recognized guidelines. Research is mostly based on randomized controlled trials (RCTs; efficacy research). Thus, examined treatment conditions, especially the treatment duration, and patients’ characteristics do not necessarily correspond to those found within routine care. Studies showed CBT packages as a whole to be efficacious, but less is known about the effects of individual CBT components. Furthermore, effects on comorbid symptoms or psychosocial impairment have been often neglected and different rater perspectives have been hardly considered in previous research. Methods This effectiveness study aimed to examine the effects of multimodal CBT in children, adolescents, and young adults (age 6–20 years) with OCD (n = 38) within routine care. Effects on obsessive–compulsive and co-existing symptoms were evaluated in a within-subject design by comparing changes during the assessment phase with 12-week standard treatment and with individually tailored extended treatment. Additionally, within the standard treatment, non-exposure treatment was compared to exposure treatment. Multi-informant assessment was applied, and the analyses included multilevel modeling and t-tests for pre-post comparisons. Results During the standard treatment and extended treatment, obsessive–compulsive symptoms, strain, and functional impairment significantly decreased. Moreover, a significant reduction of overall comorbid symptoms emerged, particularly regarding internalizing symptoms, including anxiety and depression. Comparisons of treatment components indicated that adding exposure with response prevention (ERP) has an additional positive effect. Clinical improvement and remission rates increased considerably when more treatment sessions were provided. Conclusions These results suggest that improvement after an initial 12-week course of treatment may not allow for the prediction of non-responders/non-remitters and for the termination of treatment. Overall, the findings show that results from randomized controlled trials are transferrable to routine care. Trial registration number This study was registered retrospectively at the German Clinical Trials Register (https://drks.de/search/de/trial/DRKS00030050).