Introduction

It is without a doubt that organizations are turning to the greater use of Artificial Intelligence (AI) and data analytics to drive efficiency and maintain overall competitive edge.

Internal auditors are tasked with navigating vast amount of data, and maintain precision in their assessments, challenges that heighten as the business landscape grows. AI has transformed the audit landscape, turning time consuming processes into streamlined and intelligent operations.

Most internal audit functions have embraced the shift by integrating AI in their internal audit processes leading to efficient, accurate and client-focused audit experience that goes beyond compliance to delivering strategic value.

Putting AI and data analytics into context

At its core, AI brings cognitive capability to audit processes by using machine learning (predictions of future events on the basis of past data), and natural language processing (support analysis of unstructured information like emails or contracts). AI supports internal auditing with pre-emptive vision, administrative task automation, and constant tracking of risks.

Data analytics refers to the methodical examination of data sets to locate anomalies, trends, and gaps in control. In the context of internal audit, data analytics allows auditors to transition from sample-based testing to full population analysis, thereby improving assurance and reducing the risk of missed errors or undetected fraud.

Common data analytics tools used in internal audit are;

  • Alteryx: A powerful data analytics and visualization platform that helps internal auditors understand their data. It provides a variety of tools that enable users to clean, transform, and analyze large datasets from different sources, all without requiring coding skills. Some of its key features include data blending, predictive analytics, reporting and visualization, and automation
  • Power BI / Tableau: Data visualization software which allows auditors to uncover and present data insights
  • Excel (with Power Query / Power Pivot): Allows internal auditors and analysts to perform a wide range of basic to intermediate data analytics tasks, including, data cleaning, preparation, analysis and visualization
  • Python / R: For more advanced analytics, scripting, and predictive modeling
  • ACL Robotics: Control testing, transaction risk scoring, and automation
     

The practical implementation of AI and data analytics in internal audit

  1. Detecting procurement irregularities in a manufacturing environment

    Deployment of data analytics tools supported by AI models trained on procurement behavior, allows full population testing of procurement transactions in a manufacturing company, detecting anomalies which would otherwise go undetected. This approach helps in real time detection of irregularities such as unusual reactivation of dormant vendors, split payments, mismatched invoices and purchase orders, unusual high value transactions, frequent purchase order amendments post approvals, payments processes outside working hours, duplicate invoices with slightly altered same vendor details and changes in vendor payment details. The detection of these anomalies allows the internal audit team to intervene early and recommend stronger controls.

  2. Improved audit planning through efficient analysis of multiple data points


    Conventionally, the risk assessment phase of internal audits has relied on prior period audit findings, management input and risk registers to inform on the high-risk areas. Deployment of data analytics tools and AI models allows the analysis of specific data points such as inventory, procurement and payables data to detect specific patterns, with these insights allowing the internal audit team to focus on real transactional risk. As such, the ultimate audit coverage will align with the actual risk exposure, leading to targeted and effective approach. 

  3. Analysis of master vendor data

    AI can be used to automate routine and time-consuming tasks in the internal audit process including but not limited to data extraction and processing, reconciliations and tests of controls particularly those that are transactional in nature, increasing both coverage and frequency while reducing manual efforts. In the traditional audits, the review of master vendor data is majorly a manual and heavy spreadsheet task, one that becomes increasingly complex and time-consuming for clients with large volumes of vendors and high transaction frequency. 

    By deploying data analytics tools, the internal audit team can easily review and analyze master vendor data for duplicate vendor accounts, missing key fields, unusual vendor payment details and incomplete approval trails. The automation of such tasks across the entire population reduces the review time from days to hours, enabling internal auditors to focus on investigating the root causes of the anomalies detected, and thereof recommending areas of improvement. 

  4. Continuous monitoring of procurement transactions

    For clients with high procurement volumes, the internal auditors can consider continuous monitoring through data analytics and AI models, which scans daily procurement transaction logs for any irregularities such as changes in suppliers’ details, frequent approval by the same person, and unusual purchase orders. With such, alerts shall be triggered and shared with the management for immediate follow-up. The real-time assertion system minimizes the time gap between issue identification and its remediation allowing organizations to have a dynamic audit environment where trust and transparency are continuously enhanced. 

Bottom Line

For internal audit to remain a forward-looking and value-adding function, the adoption of AI and data analytics is no longer an option, it's an imperative. The adoption of AI and data analytics allow auditors to provide more assurance, anticipate risk, and offer more powerful insights at greater speed and scale. With great power, though, comes great responsibility as internal auditors must build the competency, governance, and ethical framework to apply these tools wisely, so that technology supports rather than replace the essential principles of independence, objectivity, and professional judgment.
 


 

Caveat

This publication has been prepared by RSM (Eastern Africa) Consulting Ltd, and the views are those of the firm, independent of its directors, employees and associates. This publication is for general guidance, and does not constitute professional advice. Accordingly, RSM (Eastern Africa) Consulting Ltd, its directors, employees, associates and its agents accept no liability for the consequences of anyone acting, or refraining from acting, in reliance on the information contained herein or for any decision based on it. No part of the newsletter may be reproduced or published without prior written consent. RSM (Eastern Africa) Consulting Ltd is a member firm of RSM, a worldwide network of accounting and consulting firms. RSM does not offer professional services in its own name and each member firm of RSM is a legally separate and independent national firm.