We examined the agreement between diagnoses assigned based on the Development and Well Being Assessment (DAWBA) information collected online, and ordinary day-to-day diagnostic assignment by Child and Adolescent Mental Health Service (CAMHS) clinicians. Diagnoses were compared for 286 patients. Raw agreement for diagnostic categories was 74-90%, resulting in kappa values of 0.41-0.49. Multinomial regression models for 'emotional diagnosis' and 'hyperkinetic/conduct diagnosis' were significant (P < 0.001). Age, gender and number of informants significantly contributed to the explanation of agreement and disagreement. Agreement on mental health diagnoses may be sufficient to replace routine clinical assignment of diagnoses with an online clinical assignment, thereby saving time and resources.