New Merchants Face Elevated Fraud Risk in Their First 90 Days

New merchants are particularly vulnerable to fraud during their earliest months of trading, with fresh analysis from Fraudio showing that 10.5% of newly observed businesses experienced at least one fraudulent transaction within their first 90 days.

New Merchants Face Elevated Fraud Risk 

The study examined more than 127.5 million transactions between January 2025 and April 2026, including 3,923 merchants first seen during 2025.

Among those hit by fraud in their first three months, the median time to the first incident was just 33 days. A quarter experienced fraud within nine days, while 47% were affected within their first month.

Fraudio also found that first fraud events were heavily concentrated at the start of the merchant lifecycle, occurring at almost three times the daily rate during the first week compared with days 31 to 90.

Ordinary-Looking Transactions Can Still Be Fraudulent

The findings also highlight the limitations of assessing payments in isolation.

Fraudio found that 82% of fraudulent transactions appeared unremarkable when judged only by amount, currency and payment channel. The equivalent figure for legitimate transactions was 84%.

That suggests these attributes alone provide little separation between genuine and fraudulent activity.

Historical connections proved more informative. Across the wider dataset, 30% of fraudulent transactions involved cards previously linked to fraud, compared with just 0.075% of legitimate transactions, although Fraudio cautions that this is a retrospective measure because fraud may only be confirmed later through chargebacks.

Early-Stage Merchants Need Context From Day One

The broader risk is also materially higher. Fraudio found the fraud rate per transaction during a merchant’s first 90 days was 2.9 times higher than for payment programmes more than a year old.

For acquirers, PSPs and fraud platforms, that creates a structural challenge: the point at which merchants are most exposed is also when there is least historical behaviour available to establish what “normal” looks like.

The implication is clear, Fraud prevention for new merchants cannot rely on waiting for sufficient local transaction history to emerge. It increasingly depends on broader network context, connected-entity intelligence and shared signals from the outset.