This paper reports the psychometric characteristics of a measure that assesses the treatment outcomes of individuals with serious and persistent mental illness. Unlike other outcome measures, this one is designed to be embedded in the clinical process of planning and evaluating treatment. It collects individualized data, structures how the data are used to plan and evaluate a client's treatment, and produces aggregate information relevant for research and program purposes. Two parallel versions were developed: one for the client's self-report, and one for an informant's report. The self-report measure was administered by peer-interviewers to 244 community interviewees, and by inpatient peer-interviewers to 93 inpatient interviewees. The community interviewees also completed the BASIS-32 and SF-36. Informants for 103 of the community interviewees completed the informant version of the measure, and the CCAR. Inpatient staff completed the informant version for 161 inpatient residents without regard for matching the 93 inpatient interviewees. The two versions had acceptable internal consistency, test--retest, and interrater reliabilities. Correlations of the community interviewees' and informants' results with the BASIS-32, SF-36, and CCAR provided evidence of convergent and discriminant validity, as did contrasts between community and inpatients interviewees. The usefulness of the instrument for clinical, program and research purposes is discussed, with emphasis on the characteristics that enhance its value in clinical practice --- assessment of meaningful outcomes, operationalization of client empowerment, comprehensiveness, easy administration, and continuity across time and provider. Also discussed is a computer-based program to summarize and present the results in a rapid, clinically meaningful manner.