Helcim, a Canadian payments company that helps small businesses accept cards and manage transactions, has implemented an AI-powered risk scorecarding system to improve how it approves and monitors merchant accounts. The system is designed to proactively identify potential financial risks or fraudulent behaviour before they affect the company or its customers.
By using machine learning models trained on historical transaction data and industry risk factors, the platform can predict the likelihood of fraud or insolvency with more than 90% accuracy. These predictions help the team assess risk earlier in the onboarding process and monitor merchant activity more consistently over time. The goal is to balance speed and security — giving merchants a faster onboarding experience without compromising Helcim’s risk standards.
The project was led by a cross-functional team that brought together data scientists, product managers, and operations specialists. Together, they developed and implemented the model while building the internal processes needed to support it. A change management plan helped guide the adoption of the new tool across departments and ensure it was embedded into day-to-day workflows.
Helcim says one of the key lessons from the project was that AI works best when it complements human decision-making. While the model automates the initial identification of high-risk accounts, final decisions are still reviewed by analysts. This approach allows the operations team to focus more of their time on complex, ambiguous cases — while the model handles routine assessments consistently and at scale.
The risk scorecard also helped standardize how merchant risk is evaluated across the company. Rather than relying on subjective judgment or inconsistent criteria, the AI tool provides clear scores and insights based on reliable data. This consistency has helped Helcim streamline operations, reduce bias, and make faster, more informed decisions.
The results have been significant. Since implementation, the scorecarding system has contributed to a noticeable reduction in financial losses and an improvement in overall profit margins. Operationally, it has helped accelerate onboarding timelines, increase conversion rates, and improve the merchant experience by reducing friction during account setup.
Internally, the project also marked a cultural shift. It demonstrated how AI can drive real results when supported by strong collaboration and change management. The success of the scorecarding system has since inspired other departments at Helcim to explore how AI might improve their processes — from customer support to product personalization.
The company has also put measures in place to ensure the system continues to evolve. Designated champions from the Data, Development, and Risk Ops teams help manage ongoing improvements to the model, review its outputs, and ensure that it remains aligned with both business goals and compliance standards.
For Helcim, the introduction of AI has not been about replacing people, but empowering them. By giving teams better tools, the company has made it possible to scale smarter, serve merchants faster, and stay ahead of risk in a constantly evolving payments landscape.
This article is part of Innovation+ in the Plus 15, a special editorial series from the Calgary Innovation Peer Forum and Digital Journal that explores how Calgary-based companies are innovating.
