MSC ACCOUNTANCY

Unlocking the Power of Analytics in Audit

In an environment with increasingly complex and high-volume data, the use of technology and data analytics allows auditors to gain a deeper understanding of an entity to enhance the quality of an audit.

Using analytics improves audit efficiency by facilitating the performance of tedious tasks, freeing the auditor to focus on risk areas more effectively. The increasing use of analytics is also largely driven by stakeholders’ expectations. In many jurisdictions, including Singapore, auditors are expected to use technology, particularly data analytics, when performing audits.

From Hindsight to Foresight

The application of data analytics varies in the level of complexity: from the simplest form of historical data analyses to more complex forms involving the use of modelling tools to predict results. Parallels can be drawn with Gartner’s Four Stages of Data Analytics Maturity:

• Descriptive analytics – what has happened?
• Diagnostic analytics – why did it happen?
• Predictive analytics – what will happen?
• Prescriptive analytics – how can we make it happen?

The current application of analytics in audit focuses on the descriptive and diagnostic stages, analysing historical data to identify outliers, and providing hindsight into past activities. In most audits that adopt analytics, the use of visualisations with dashboards and interactive infographics with drill-down features have become common. Major development in computing capabilities and storage capacities have made this possible.

However, by exploring the more complex realms of predictive and prescriptive analytics, which provide actionable insights and foresight, there is tremendous potential to further enhance audit quality. Exciting advancements in the field include regression analysis to predict and identify trends, patterns and relationships, and the use of complex modelling tools such as machine learning, cognitive computing, and artificial intelligence to perform predictive analytics and clustering analysis.

Seeing the Bigger Picture

Let’s look at how three commonly used audit analytics can provide a clearer picture of operations and potential risk areas. The figure below illustrates general ledger analysis using visualisation tools on a company that sells copiers.

The use of visualisation allows the auditor to immediately see the fluctuating and increasing sales of Copier D as depicted in the yellow arrow. Why did sales decline from 2018 to 2019, then peak again in 2020 to become the highest contributor among all copiers? Why were such fluctuations only peculiar to Copier D? Is Copier D a seasonal product or was there a special marketing promotion that could explain the change? Or is there a risk that the revenue and associated expenses were not recognised in the correct period?

Visualisation tools are also commonly used for journal entry testing. Historically, there have been numerous accounting scandals, such as WorldCom, that involve management overrides of controls around the journal entry process. To address this risk, auditing standards mandate that the auditor tests the appropriateness of journal entries. Because of the high volume of journal entries, visualisation tools enable the auditor to focus its energy on the highest-risk journal entries culled from a full set, rather than testing a random sample.

The ability to test whole data populations through automation also increases the efficiency of a revenue three-way match. The figure below illustrates a typical outcome of the digital matching of all sales invoices, shipping documents and sales orders of an entity with $130 million in revenue generated from 850,000 transactions.

Digital matching of all the sales invoices, shipping documents and sales orders of an entity with revenue of $130 million generated by 850,000 transactions:

Most of the prices and quantities are matches. However, there remain a number of mismatches – specifically, 15,150 transactions where the quantities do not match and 23,250 transactions where the prices do not match. These outliers could represent fraud or errors, or there could be genuine reasons behind the differences, which will require the auditor to investigate and resolve.

Ultimately, while automation undeniably increases audit efficiency, audit effectiveness remains dependent on the integrity of data used, how the auditor follows up on identified outliers, and the reliability of the analytics tools and evidence gathered.

Preparing for the Future of Auditing

There remain significant areas that the auditing profession and standards setters are working hard to resolve before we can expect analytics to be widely used in the audit of financial statements. Even so, the prospect that analytics can transform how we traditionally perform an audit is exciting. There is the possibility that in future, data volume may exceed human capacity to audit, which makes the application of analytics a must-have for every big audit engagement.

At Nanyang Business School, our Data Analytics and Machine Learning module imparts key concepts of predictive analytics in accounting. Students will have hands-on experience learning to build analytical models using real-life case studies. The integration of data analytics courses into our MSc Accountancy curriculum is a critical aspect of preparing our students for the ever-changing digital world.

Goh Kia Hong is a senior lecturer at Nanyang Business School, Nanyang Technological University. He teaches audit and risk management for the bachelor and master’s degree programmes.

Talk to us to find out more about the MSc Accountancy programme. Click here to book an appointment today.

Download MSc Accountancy Brochure