Objectives The present study extracted symptom profiles based on parent and youth report on a broad symptom checklist. Profiles based on parent‐reported symptoms were compared to those based on adolescent self‐report to clarify discrepancies. Method The current study used archival data from 1,269 youth and parent dyads whose youth received services at a community mental health center. The mean age of the sample was 14.31 years (standard deviation = 1.98), and the youth sample was half male (50.1%) and primarily Caucasian (86.8%). Latent profile analysis was used to extract models based on parent and self‐reported emotional and behavioral problems. Results Results indicated that a 5‐class solution was the best fitting model for youth‐reported symptoms and an adequate fit for parent‐reported symptoms. For 46.5% of the sample, class membership matched for both parent and youth. Conclusion Latent profile analysis provides an alternative method for exploring transdiagnostic subgroups within clinic‐referred samples.