United KingdomTechnology and telecom2025 – ongoingFresha
Machine learning for payment fraud detection
Fresha reduced fraudulent transactions by 80% with machine learning
Problem
Fresha's payments platform faced stolen cards, fake businesses and account takeovers. Its fraud analysts needed additional tools as the business grew.
Approach
Fresha brought in data scientists to develop fraud detection models using multiple signals. The system checked transactions automatically and queued suspicious items for manual review, using transaction data refreshed in Snowflake.
Results
Takeaway
Fresha's 80% fraud reduction came from an ML workflow that retained manual review of suspicious transactions.
Categories
- Industry
- Technology and telecom
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Goal
- Quality and risk
- Evidence
- Vendor case study
Sources
Source summaries
Snowflake reported that Fresha's ML models flagged suspicious payments for human review. Fraud fell by 80% when the approach began. The page carries no date; the Internet Archive first saved it on January 16, 2025.
Full details are in the original sources.
