The Bank of Japan published a Review Series article outlining how it uses supervisory granular data, particularly loan-by-loan data collected from banks, in its recent financial system analysis. The paper argues that transaction-level data can reveal facts that are difficult to detect from conventional aggregated data, supporting earlier identification of vulnerabilities and assessment of potential risks. The authors highlight the need to continue improving analytical methods for using granular datasets to strengthen financial stability monitoring and contribute to better risk management practices at banks. The report is published as an English translation of the Japanese original and reflects the views of the authors rather than the Bank of Japan.