From Training to Performance: Evaluating Digital Skill Transfer in Commercial Banks of Uzbekistan

Digital banking creates value only when employees can convert newly acquired knowledge into reliable work behavior and measurable operational outcomes. This study develops a reproducible framework for evaluating the transfer of digital skills training to job performance in commercial banks. The research combines human capital theory, training-transfer theory, task-functional analysis, and a review of official banking and regulatory documents in Uzbekistan. The proposed architecture separates five levels of evidence: investment inputs, competence acquisition, workplace transfer, labor outcomes, and economic outcomes. It introduces a transfer-adjusted competence effect, a difference-in-differences estimator for labor indicators, and a conservative procedure for calculating full training investment and monetizable benefits. The framework also specifies measurement timing, comparison-group selection, data sources, and attribution-confidence grades. A pilot protocol is designed for two organizationally different banks, with employee-level results to be collected only through anonymized baseline and follow-up assessments. The methodological results show that course completion and test scores are insufficient evidence of productive human capital. The decisive link is workplace transfer, which requires a preassigned work task, access to the relevant digital system, managerial support, and traceable performance data. The framework enables banks to distinguish successful learning from successful application and to renew training budgets on the basis of verified labor effects rather than participation counts.