Data Analytics Automation: How Public Sector Companies Can Take Controls Testing to the Next Level

Government and public sector organizations today are under enormous pressure. They are constantly working to generate more value, enhance their operational efficiencies, and attract and retain skilled employees—all while facing increasing levels of risk and public scrutiny. 

Within this challenging environment, the importance of having meaningful governance and effective internal controls has never been greater. The challenge is that managing and monitoring internal controls is a complex, time-consuming, and highly manual exercise. 

In 2019, Guidehouse partnered with the Association of Government Accountants on an intelligent automation survey. A large majority of respondents to this survey saw considerable potential for the use of automation in the government audit process—with 83% saying it could improve data accuracy, 80% saying it could improve data availability, 60% suggesting it could improve security of information systems by reducing human error, and 50% believing it could reduce the complexity and cost of audits.  

While many controls testing groups within government and public sector organizations see enormous potential for data analytics automation to help them enhance their management of internal controls, far fewer have moved forward with integrating automation into their activities. This is because it can be very difficult for controls-testing groups to determine how to move forward with integrating data analytics automation within their activities at a level that meets assurance standards.

In this position paper, we explain what data analytics automation is and examine the key factors driving its use before introducing our four-step approach to helping organizations more effectively integrate data analytics automation into their controls-testing processes. Whether you are new to the concept of using data analytics automation or want to enhance your approach, this paper will give you a strong place to start.

 This paper is co-authored by Farhan Bandukda.