Arming accountants with fraud detection software in the war on corporate crime

Engine B - March 17th, 2021

Fraud detection software

This week, the Institute of Chartered Accountants for England and Wales (ICAEW) published an article championing the role of both accountants and AI technology in the war against economic crime. It’s no secret that the audit sector has been rocked with financial scandals recently.

And, today, on Thursday, March the 18th 2021, the much-anticipated whitepaper from Business Secretary, Kwasi Kwarteng on audit reform reinforces this message and highlights the urgent need for change: “It’s clear from large-scale collapses like Thomas Cook, Carillion and BHS that Britain’s audit regime needs to be modernised with a package of sensible, proportionate reforms.”

Engine B fully supports these audit reforms and that there is ‘a huge overlap between the core skills of accountants and what it takes to fight financial crime’. The sector urgently needs to equip auditors with the tools to enable them to detect financial fraud quicker, and of course, avoid future scandal.

As the International Federation of Accountants (IFAC) Director of Public Policy and Regulation, Scott Hanson, suggests in the ICAEW article, the profession is moving in the right direction. And the publication of the government’s long-awaited report is a huge step forward.

But how can auditors and fraud detection software truly work in harmony to deliver reliable audits and avoid the glare of future newspaper headlines? And what specific technologies should audit look towards to build back public trust and confidence in the sector?

Fraud detection softwareA new approach is needed

 As we’ve seen, financial frauds are expensive and fraudsters are continually finding new ways to exploit monitoring systems. The problem lies in the current approach to identifying fraud.

Many audit firms try to tackle fraud by analysing transactions in isolation. This method of fraud analyses has its limitations and prevents auditors from seeing the bigger picture. Audit firms need to adopt a context-driven view of information, one that uncovers connections across silos of information so data can be analysed in a meaningful and more intelligent way.

Over the past few decades, the exponential rise of machine learning (ML), the introduction of Deep Learning Techniques and the dramatic improvement in the analytical capabilities of software are enabling improved anomaly detection. Yet, while anomaly detection technology has been utilised in some aspects of audit for a few years, it’s not often applied in a way that allows it to be used within a high-quality methodology.

Fraud detection software as an enabler Fraud detection software

Technology alone is not enough. Auditors should consider fraud detection software more as an enabler to better insights and decision-making rather than a substitute for professional opinion. Of course, an auditor’s professional advice is something that cannot, nor should ever, be replaced by a machine. However, by combining the auditor’s expert judgements and leveraging fraud detection software to uncover hidden risks and anomalies faster, both audit firms and their clients can be confident that potentially fraudulent activity will not be overlooked.

Fraud detection using Knowledge Graphs and Machine Learning

Engine B’s Fraud Detection Knowledge Graphs allow auditors to interrogate and analyse their client’s data in a powerful, context-driven way by detecting data abnormalities or problematic records and intrusion detection. While auditors know what kind of market trends and management incentives create a risk of fraud, it’s much harder to know where to look for specific vulnerabilities or evidence.

Using Machine Learning and Knowledge Graphs for fraud detection makes this easier by identifying hidden anomalies in corporate data and making these anomalies highly visible to allow the auditor to see exactly what looks suspicious and how it relates to its context.

All of the evidence is synthesised and presented together with explanations for what’s normal and what’s not making assessments based on risk far more reliable.

Fraud detection software

Engine B’s Fraud Detection Knowledge Graphs also vastly reduce the time it takes to investigate fraud and frees up time for firms to adopt a robust ISA 315 process, clearly document risks and planned responses and focus their controls and substantive testing where it’s most needed.

The result is that firms are much less likely to be caught out by fraud or error hidden deep in the detail of millions of complex transactions that look to the human eye like normal business behaviour and auditors can identify instances of fraud that would otherwise not been possible to see in traditional analyses tools.

The future of fraud detection

Engine B’s fraud detection software gives auditors a laser-like focus on risk. Sure, fraudsters will always be looking for new ways to innovate, take risks and exploit new avenues. But when the auditor leverages fraud detection software, they receive a rich view of any unusual behaviour and the context of that behaviour. This makes it easier for them to see what’s relevant, to apply their judgement to the risk, and subsequently, to detect fraud before it’s too late.

The conclusion? Technology really is the auditor’s ally in the fight against corporate crime.

Find out more how your firm can utilise Engine B’s Knowledge Graphs for fraud detection by downloading our guide.

Share this...
Share on LinkedIn
Linkedin
Tweet about this on Twitter
Twitter