Fraud Lives in the Gaps Between Processes. Technology Can Close Many of Them

Look closely at how most frauds worked and a pattern appears. The fraudster did not break a control. They used a space between two of them. Procurement assumed finance would check the vendor. Finance assumed procurement had. The sales team agreed a scheme by email; accounts paid the claim because it came with the right form.

The ACFE's Occupational Fraud 2026 report found that a lack of internal controls was the primary weakness in 33% of cases. Many of those "missing" controls are not missing from any single process. They are missing from the joins.

Where the gaps open

From our investigations, the same gaps recur across sectors:

• Hand-offs between functions. Vendor creation, ordering, receipt and payment sit with different teams, and nobody sees the whole chain.
• Work outside the system. Spreadsheets for scheme calculations, rebates and incentives; approvals by email or messaging apps; side letters that never reach the ERP.
• Master data nobody owns. Vendor, customer and bank details changed by whoever has access, with no independent verification.
• Manual workarounds that became permanent. A temporary bypass during a system issue that was never switched off.
• Data that is never connected. HR records, vendor master data and payment data held in separate systems, so an employee's address on a vendor record goes unnoticed.
• Migration and change windows. New systems going live, acquisitions being integrated, or processes being outsourced, when controls are relaxed and data moves by hand.

How technology can close them

Technology is most effective against fraud when it does one of three things: removes discretion, connects data or records exceptions automatically.

Remove discretion

• System-enforced workflows for vendor onboarding, purchase orders and payments, with approval limits the user cannot change.
• Automated three-way matching of order, receipt and invoice, with exceptions blocked rather than flagged.
• Blocked self-approval, so no user can approve a transaction they created.

Connect data

• Master data verification against independent sources, including call-back checks on bank account changes.
• Cross-system matching of vendor, customer and distributor records against employee and director data.
• Continuous transaction monitoring, running scheme-based tests on payments, journals, credit notes and claims as they happen. Our view on hypothesis-led analytics explains how to design them.

Record exceptions

• Automatic logging of overrides, manual journals and configuration changes.
• Privileged access management, so administrator activity is recorded and reviewed by someone independent.
• Digital audit trails for approvals that previously happened by email or message.

What good looks like

1. Map before you buy. Walk through each high-risk process end to end, including the informal steps, and mark every hand-off, spreadsheet and workaround.
2. Rank the gaps by the schemes they enable, drawing on your fraud risk assessment.
3. Configure what you own. Many ERPs already support approval workflows, matching and logging that have been switched off or never set up.
4. Add tools where they close a specific gap, not because a category of product exists.
5. Assess every change for fraud risk before go-live, including migrations, integrations and outsourcing.
6. Test that it works with data, months after implementation.

Common mistakes

• Automating a broken process. A faster bad process loses money faster.
• Trusting the system blindly. Configuration can be changed; who can change it, and who checks, matters.
• Leaving informal channels in place. If approvals still happen by message, the workflow is decoration.
• Ignoring the people side. Technology that makes legitimate work harder creates new workarounds.
• Buying dashboards nobody reads. Exceptions need owners.

How we can help

Technology choices are easier with an independent view of where the gaps are. Our consulting work maps high-risk processes from a fraud perspective, reviews how existing systems are configured and used, identifies the gaps that matter, and helps companies decide which technology will close them, whether by reconfiguring what they have or adopting something new. Our forensic accounting and data analytics team builds the monitoring tests, and our digital forensics team advises on logging and evidence readiness. If you are planning a system change, or wondering whether your systems are doing what you think, speak to us.