The research was conducted with the aim of designing a smart educational leadership model to improve school principals' decision-making. The research method is applied in terms of its purpose and is data-driven theory (systematic design) in terms of data type. The statistical population of the research was school principals in Tehran. Sampling was conducted using purposeful and snowball methods. The samples were selected according to the course of theory formation and based on the theoretical saturation rule, with a total of 19 people. The research data collection tool was a researcher-made semi-structured interview. The data were analyzed using Max Quda software (version 2020) in three stages of open, axial, and selective coding, and its reliability was confirmed through agreement between the coders. From the data analysis, 61 open codes, 14 axial concepts, and 6 main categories were extracted in a paradigmatic framework. The final model showed that the central phenomenon of "intelligent educational leadership" is manifested in three interconnected dimensions: strategic rationality (including data-driven decision-making, systematic analysis, and predictive thinking), emotional-communication intelligence (including relationship management, empathy, and trust-building), and spiritual-moral orientation (including motivation, adherence to professional ethics, and student centeredness). This phenomenon is formed under the influence of causal conditions (internal and external transformational pressures), context (intraorganizational and extra-organizational factors), and intervening conditions (structural facilitators and inhibitors), and through action strategies (including core competencies and executive-process measures), it leads to desirable outcomes at two process levels (improving the quality, speed, and coherence of decision-making) and organizational levels (improving educational performance, stakeholder satisfaction, and transforming into a learning organization).