Purpose The purpose of this study is to derive the predictive variables for employee proactive behavior by developing a predictive model for employee proactive behavior. Methods The data utilized in this study were obtained from the second and third waves of the Human Capital Corporate Panel (HCCP), provided by the Korea Research Institute for Vocational Education & Training, comprising 5,588 observations. Using the random forest technique, a machine learning method, 101 organizational-related variables were applied to the research model to predict proactive behavior. Results We developed the predictive model for employee proactive behavior, which has 74.30% accuracy in model performance. Furthermore, by focusing on the top 30 important variables, we analyzed the primary variables predicting employee proactive behavior by comparing the key characteristics. Conclusion Based on the research findings, implications for nurturing and managing proactive talents and behaviors, which are essential for sustainable organizational performance from a human resource management perspective, were discussed.