Formulating treatment goals has been shown to be an area of vital concern for both outcome and treatment processes. However, it is not as yet an area of structured routine assessment, either in clinical practice or in research. One possible explanation for this is the lack of validated and readily available goal assessment procedures. The present study investigated the test-retest stability of a Swedish translation of the checklist version of the Bern Inventory of Treatment Goals (BIT-C) among 30 patients in primary care. We calculated the consistency of the endorsement of the different therapeutic goal categories over a 2-week period prior to treatment. There were no changes in symptoms or quality-of-life-related measures between the two assessment points. Overall, the goal category items in BIT-C were found to demonstrate moderate to substantial reliability. In conclusion, even though our study was small, it provided initial psychometric support for the Swedish version of BIT-C as a clinically useful tool for the assessment of treatment goals.