An auditor can view themselves as many different personas, but up until recently ‘audit data analyst’ was not one of those personas. The truth is, I’ve always thought that this was a bit of an unfair position for auditors.
For as long as I have been involved in the accounting and finance industries, auditors have been drawing conclusions about large populations of data by using random sampling or a particular strategic lens. What has always impressed me is how a seasoned partner can spot an error deep in the numbers just by looking at the primary statements.
While strong audit data analysts are still applying their incredible talents, many auditors are beginning to leverage new audit technologies to streamline their analysis methods.
Embracing new data analysis techniques during audits
What’s most interesting today is how professional data analytics techniques from other fields are being combined with traditional audit approaches. This has enabled new ways for auditors to interrogate, understand, and gain assurance during data journal entry analysis or general ledger analysis. This ranges from basic aggregation techniques such as calculating proof in totals and creating moderately complex data visualizations to machine learning techniques designed to spot unusual patterns.
Using AI-powered technology such as Ai Auditor, audit data analysis appears to be entering a new phase of progression. AI audit solutions leverage machine learning to analyze general ledgers and deliver automated risk scores across all transactions and financial data.
How the role of the data analyst is evolving with AI technology
Learning how to properly implement these technologies to evolve auditing processes and general ledger analysis requires consideration. However, I have seen many instances where these cutting-edge audit analysis technologies were able to flag truly interesting items such as the purchase of a Porsche for a company director. When one experiences these types of results with AI audit software, it’s easy to believe that the future is here for journal entry analysis. And, long gone is the day of manual data segmentation in Excel.
Many of these AI audit solutions work by building some expectation of normal within a specific pool of data. The many breakthroughs that are still occurring in data science and artificial intelligence will likely improve the machine’s sense of nuance. As more accurate models involve higher levels of complex analysis, we must, as an industry, weigh this fact against our need for explainable results.
This is not the end for analyzing audit data. Some auditors will always carry the persona of data analysts because they are inherently great at decoding data. However, perhaps that role is evolving alongside new AI audit technology. And perhaps, that’s a good thing.