In this study, we have developed a process to construct microstructure-micromechanical property correlations based on 3D visualization. The 3D microstructure reconstruction is accomplished through serial sectioning using a RoboMET.3D, which facilitates automated and controlled removal of material layers and uses an integrated inverted optical microscope for imaging. Image stacks obtained from the serial sectioning undergo a process involving phase segmentation, correlation analysis and feature extraction. Two-point statistics and principle component analysis record important information contained in the microstructure, which is used as input for regression modelling. Nanoindentation mapping is employed to determine hardness and Young’s modulus. Multiple regression methods are then used to predict micromechanical properties for different characteristic microstructures. The best performing model is applied to another image stack to achieve 3D property reconstruction. Our approach establishes a comprehensive microstructure-property correlation through 3D visualization, providing new insights into the mechanical behavior of pearlitic steels. This approach holds the potential to predict additional microstructure-related properties and can be extended to other material systems.