Safety & Security

62% of finance teams hit by fraud in three years, SAP Concur finds

SAP Concur's Finance Leaders' Fraud Report 2026 finds 62% of finance teams lost money to fraud in three years; only 22% feel very well protected.

SAP Concur warns of rising fraud losses in finance
SAP Concur warns of rising fraud losses in financeAI-generated
By James Calloway6 min read

Why it matters

  • 62% of surveyed finance and AP professionals reported fraud losses in the past three years; only 22% feel very well protected.
  • Impersonation fraud rose from 35% in 2025 to 40% in 2026; business email compromise was cited by 45% of respondents.
  • 6% of respondents reported deepfake audio fraud and 6% deepfake video fraud.
  • 29% of organisations invest nothing in fraud training; 76% of respondents want more training and resources.
  • 60% of respondents believe their manager could do more to protect their team from AP fraud.

Sixty-two per cent of finance and accounts payable professionals say their organisation has suffered financial losses from fraud in the past three years, according to SAP Concur's Finance Leaders' Fraud Report 2026. The same global survey found that only 22 per cent of respondents feel very well protected against fraud. That gap between exposure and confidence is the central finding of the report, published on 20 November 2026, and it frames a problem that finance functions cannot solve with approval workflows alone.

The stakes are straightforward. Finance and AP (accounts payable, the function responsible for paying suppliers and processing invoices) teams sit at the point where money leaves the business, and the report suggests the criminals targeting them are getting better at mimicking legitimate requests. The survey draws a picture of teams that know they are exposed, doubt their defences, and are not consistently trained to close the gap.

Which fraud methods are hitting finance teams hardest?

Business email compromise led the list of reported incidents, cited by 45 per cent of respondents. Impersonation fraud followed at 40 per cent. The report also documents the arrival of manipulated media in the finance fraud toolkit: six per cent of respondents reported deepfake audio fraud and another six per cent reported deepfake video fraud.

The impersonation figure deserves particular attention because it is moving. According to the research, impersonation fraud rose from 35 per cent in 2025 to 40 per cent in 2026 — a five-point increase in a single year. That trend puts direct pressure on the finance staff responsible for checking payment requests, supplier details and other transactions, because impersonation attacks are specifically designed to mimic legitimate requests or internal approvals.

The deepfake numbers are small, and the report does not claim they represent a majority threat today. But their appearance in the incident data marks a broader shift in fraud risk: from conventional payment scams toward more convincing forms of impersonation and manipulated media. For companies, the report notes, that shift raises operational as well as technical questions — who approves payments, how exceptions are handled, and what information gets checked before money leaves the business.

The takeaway in plain words: the most common attacks still run through email and impersonation, but audio and video manipulation has entered the incident record.

What does SAP Concur say finance leaders should do?

Fabian Calle, Managing Director of SAP Concur Australia and New Zealand, argues that traditional controls have been outpaced. "Fraud is becoming harder to detect as criminals use increasingly sophisticated methods to exploit weaknesses in financial processes," Calle said. "Finance leaders can no longer rely on traditional approval workflows and manual checks alone. They need greater visibility into spending, stronger controls, and the ability to identify suspicious activity before it results in financial losses."

Calle also flagged the resource problem that smaller organisations face. "Businesses of all sizes face these challenges, though smaller finance teams can be particularly vulnerable when limited resources make it difficult to review every transaction," he said. "Strengthening fraud resilience starts with understanding where risks exist and putting practical safeguards in place."

The report identifies artificial intelligence as one tool finance teams can use to improve oversight. In practice, that means systems that flag unusual spending patterns, duplicate invoices and other anomalies across travel, expense and invoice processes. SAP Concur, it should be noted, sells exactly this category of software, so its enthusiasm for automated oversight carries an obvious commercial interest. The report itself concedes the limits of the pitch: its findings suggest technology alone is not seen as enough.

Calle makes the same point directly. "AI can help finance teams identify potential fraud earlier, though technology alone isn't enough," he said. "Organisations need to combine automated controls with clear policies, employee training, and human judgement. The goal is to make suspicious activity easier to identify without creating unnecessary administrative work."

The stated benefit of the automated approach is that it reduces reliance on manual checks and lets staff focus on higher-risk transactions. That is a efficiency claim as much as a security one, and the report offers no benchmark data on how much fraud such systems actually prevent.

Can connected data actually catch fraud?

Calle argues that joining travel, expense and invoice data into a single view is what gives finance leaders the ability to see irregularities at all. "Connecting travel, expense, and invoice data also gives finance leaders a clearer picture of where money is going," he said. "This visibility helps them identify irregularities, strengthen compliance, and make more informed decisions about financial risk."

The logic is that fraud hides in the seams between systems — a duplicate invoice here, an unusual expense pattern there — and that unconnected data lets it stay hidden. The report does not provide case studies or quantified outcomes from organisations that have adopted this approach, so the claim rests on the survey's exposure data rather than on demonstrated results.

In plain words: the vendor says connected data finds irregularities, but the report shows the problem, not the proof.

How big is the training gap?

The survey's training findings are arguably its most damning. Sixty per cent of respondents said they believe their manager could do more to protect their team from AP fraud. Nearly three in 10 — 29 per cent — said their organisation does not invest in fraud training or awareness programmes at all. At the same time, 76 per cent said they would welcome additional training and resources.

Those numbers sit awkwardly beside the 62 per cent loss figure. Organisations are being defrauded at high rates while, in nearly a third of cases, spending nothing on teaching staff how to recognise the attacks. The demand for training is clearly there; the supply, according to the respondents, is not.

The report points to regular policy reviews, supplier verification and employee education as the areas where businesses can tighten controls. Smaller finance teams may face particular difficulties here, because limited resources reduce the scrutiny given to each payment or invoice — exactly the weakness that impersonation tactics are built to exploit.

Together, the report concludes, these figures suggest fraud prevention is still treated inconsistently across organisations, despite the frequency of reported losses. In plain words: most teams are losing money, most want more training, and nearly a third get none.

What is the so-what for finance leaders?

Calle closes with a synthesis that mirrors the report's structure: people, processes and technology have to work together. "Fraud resilience depends on people, processes, and technology working together," he said. "Finance leaders who invest in visibility, automation, and employee awareness will be better positioned to manage emerging risks while maintaining operational efficiency. Taking a proactive approach to fraud prevention can help organisations protect business resources, strengthen trust, and support sustainable growth."

The caveats are worth holding onto. This is vendor-sponsored research from a company that sells expense and invoice automation, the AI recommendations align neatly with its product line, and the report offers no methodology sample size or independent verification of the loss figures. What it does establish, on the respondents' own accounts, is that fraud losses are widespread, confidence in defences is low, impersonation is rising year on year, and deepfakes have registered in the incident data. For finance leaders, the immediate questions the report leaves open are operational ones: who approves payments, which exceptions get escalated, and what gets verified before money moves.

Source: itbrief.com.au

Share this article:

More from James Calloway

James Calloway

Show full bio

News editor covering people, launches and day-to-day practice in artificial intelligence and machine learning at AI In Context.

791 articles

Related articles

  1. Adactin maps six fraud defences for Australian businesses facing AI scams
  2. Anomali warns AI agents will run entire fraud campaigns against Australian firms in 2027

« Previous articleNext article »