Purpose – The aims of the present study were to test the predictive validity of the Swedish version of the Team Diagnostic Survey (TDS). Design/methodology/approach – A model with both performance and satisfaction was tested with structural equation model (SEM) analyses. Participants completing the survey were employees (N, 214) across three large workplaces. Analyses were done at the group level and data from 33 teams were included in the final data material. Findings – Results from validation data indicate that the TDS has satisfactory high Cronbach’s alpha values on most factors. Results from the SEM analyses show a moderate model fit for the main model. Team-level factors predict both performance and satisfaction, while organization and coaching factors do not. Research limitations/implications – The present study was limited to a cross-sectional design, but earlier studies have shown that the accuracy of the TDS remains consistent over time. The main purpose of this study was to test the predictive validity of the instrument. Theoretical implications of the study are that a survey can be used to get a valid overall picture of the real-life work team’s effectiveness. Practical implications – Practical implications of the study are that communication between researcher and/or consultant and organizational stakeholders is made easier, as the most important factors that affect team effectiveness are identified. Originality/value – The work on the development on TDS has shown that it is possible to use a complex instrument to diagnose work groups, and this line of research is leading the way for better instruments. © Emerald Group Publishing Limited.