The challenge in developing molecule-based electronic materials lies in the uncontrollable or unpredictable nature of their crystal structures, which are crucial for determining both electrical properties and thin-film formability. This review summarizes the findings of a research project focused on the systematic development of crystalline organic semiconductors (OSCs) and organic ferroelectrics by integrating experimental, computational, and data sciences. The key outcomes are as follows: 1) Data Science: We developed a method to identify promising materials from crystal structure databases, leading to the discovery of unique molecule-based ferroelectrics. 2) Computational Science: The origin of high layered crystallinity in π-core – alkyl-chain-linked molecules was clarified based on intermolecular interaction calculations. We proposed a stepwise structure optimization method tailored for layered OSCs. 3) Material Development: We developed various alkylated layered OSCs, which exhibit high mobility, heat resistance, and solubility. We discovered several unique phenomena, including frozen liquid crystal phases, significant polar/antipolar control, and phase control through mixing, leveraging the variability of alkyl chain length. We also developed molecule-based ferroelectrics showing peculiar ferroelectricity, including multiple polarization reversal, competing ferroelectric/antiferroelectric order, and spinner-type configurations with π-skeletons. 4) Advanced Structural Analysis: By combining cryo-electron microscopy and X-ray-free electron laser (XFEL), we enabled crystal structure analysis for ultrathin crystals that are usually difficult to analyse. 5) Device Development: Utilizing the self-organized growth of layered OSCs through solution processes, we developed a method to produce exceptionally clean semiconductor – insulator interfaces, achieving field-effect transistors that show sharp (near theoretical limit) and stable switching at low voltages.