MODERNIZING FINANCIAL STATEMENT AUDITS THROUGH DATA ANALYTICS AND CONTINUOUS AUDITING MODELS: EVIDENCE FROM ERP-BASED ACCOUNTING SYSTEMS

Abstract

The exponential growth of digital accounting data has rendered traditional, sample-based financial statement auditing obsolete. This study investigates the integration of Data Analytics (DA) and Continuous Auditing (CA) models to modernize the substantive testing of accounting records. Addressing the limitations of retrospective auditing, this paper develops an automated Journal Entry (JE) testing framework aligned with International Standard on Auditing (ISA) 240. Utilizing a simulated General Ledger (GL) dataset containing 500,000 transactions, the methodology applies Benford’s Law and algorithmic rule-based scripts to test 100% of the accounting population. The results indicate that DA reduces manual substantive testing time by 60% while increasing the detection rate of material accounting anomalies (e.g., unauthorized weekend entries and threshold-evading disbursements) by 85% compared to traditional random sampling. The study concludes that transitioning from manual vouching to algorithmic assurance shifts the auditor’s role from mechanical data gathering to high-level financial judgment. Policy implications urge audit standard-setters to revise audit evidence guidelines and recommend that audit firms mandate hybrid accounting-data science competencies.