Combatting fraud through analytics

Fraud can often go undetected for months, if not years, resulting in high value and repeated offences. Combatting fraud through data analytics is an increasingly effective method. The Association of Certified Fraud Examiners’ most recent ‘Report to the Nations’, stated that 35% of Asia-Pacific fraud cases surveyed were detected through audit or analytical methods. More and more, organisations are proactively leveraging data analytic tools and expertise to identify, minimise, and prevent fraud.

Why is fraud increasing?

• Global economic events: The pandemic, supply chain stress, and cost of living pressures have had a significant impact on people and organisations. 

• Personal circumstances: Fraud is an act often motivated by personal circumstances, leading good people to make bad decisions. Individuals rationalise the decision to commit fraud against their employers driven by redundancies or terminations coupled with personal financial pressures.

• Resourcing issues: Organisations in some sectors have been forced to cut costs which has impacted staff numbers and the effectiveness of internal controls.

• Remote working: Organisations are more exposed to risks that they would otherwise have been able to avoid or mitigate through active physical scrutiny. With an uptick in evolved, flexible workplaces, frauds are increasingly being perpetrated remotely.

Managing fraud risk through analytics

Data analytic testing is one pillar of prevention but should not be treated as a set and forget approach. When embedded in an organisation’s broad fraud risk framework, data analytics can significantly reduce the risk of fraud occurring. A more complete approach to managing fraud risk should include:

• Discreet analytic programs examining payment types, regularity, amounts and beneficiaries that can highlight anomalous or unusual transactional activity requiring deeper review. 

• Transactions or relationships that seem irregular can often be identified through layers of testing that analyse conflicts of interest and/or gifts and benefits registers overlaid against Accounts Payable patterns to identify conflicts of interest or undisclosed relationships. 

• Vendor Masterfile data that identifies address, bank account or contact details of vendors overlaid with employee data can also identify undisclosed relationships or conflicts of interest.

• Overlaying training data with payroll data to identify gaps that indicate an employee is getting paid but has never attended or completed a training course. Could this be a ghost employee?

Business considerations to combat fraud

A comprehensive analytics program is a powerful detection tool, however organisations also should consider other preventative measures to mitigate fraud, such as:

• Assessing fraud risk by conducting a fraud risk assessment. Identify high-risk functions within your organisation and stress-test their workflows and processes to ensure controls are working.

• Refreshing and resetting system-based delegations and approvals. If a person doesn’t require system access for approvals remove it immediately. 

• Conducting background checks on prospective employees, especially those with intended financial delegations.

• Increasing employee communication to ensure organisational policies, procedures and expectations are known. Messaging should highlight the consequences of fraud.

In the absence of proactive measures and effective data tools, organisations leave themselves exposed to fraud risks, reputational damage, and potential business disruption. The cost and impact of which could be felt for longer than the fraud itself.