SaaS Dashboard
Project type
B2B Web SaaS Platform (Risk & Payments Operations)
My role
UI / UX Designer
Tools
Figma • Figjam • Shadcn UI • Confluence
Contribution
Competitor research, Product architecture, User workflows, and data dashboards.
Project overview
This product is an AI-powered fraud detection platform designed to help payment and risk teams automatically analyze transactions, detect suspicious behavior, and make faster, more accurate decisions.
The system uses machine learning models to score and classify transactions in real time. Low-risk transactions are automatically approved, high-risk ones are blocked, and only edge cases are escalated for manual review.
I designed a centralized investigation workspace that blends AI automation with human oversight, allowing analysts to focus only on the transactions that truly require attention.
The Problem
Fraud detection platforms are powerful but often difficult to use. Through competitor research, I found that many existing tools suffer from overly complex interfaces and poor usability. Risk analysts are often presented with too many configuration options, dense dashboards, and fragmented workflows. Important actions and insights frequently become buried within multiple layers of navigation, slowing down investigations.
Solution
To address the complexity and usability challenges identified in existing fraud detection platforms, I designed a streamlined, task-oriented experience centered around how analysts actually work.
The solution focused on simplifying decision-making, reducing cognitive load, and centralizing workflows into a single, cohesive interface.
Competitor Research & User Journey
To better understand how fraud analysts work with existing tools, I conducted competitor research on fraud detection platforms. The analysis revealed that many products offer powerful capabilities but suffer from high interface complexity and usability issues.
Analysts are often presented with too many configuration options, fragmented navigation, and hidden critical information, making investigations slower and more cognitively demanding.
Based on this research, I created a user journey map of an analyst using a competitor tool. The journey highlighted key friction points where users struggle to locate important information, navigate between multiple screens, and make confident decisions.
This helped identify opportunities to simplify workflows and centralize investigation tasks.
Product Information Architecture
After identifying the usability challenges in existing tools, I defined the information architecture of the platform.
The goal was to organize the product around the core tasks fraud analysts perform daily, ensuring that the most critical actions are easy to access.
The platform was structured around three primary areas:
Fraud Monitoring – dashboards and analytics for transaction activity
Orders Workspace – reviewing flagged transactions and customer profiles
Firewall / Rules Engine – configuring fraud prevention rules
This structure helped create a clear navigation hierarchy and simplified investigation workflow.
User Flows
Once the product structure was defined, I mapped the core user flows to understand how analysts would interact with the system during investigations.
The focus was on simplifying the investigation process and reducing the number of steps required to make a decision.
Design System & UI Architecture
To ensure consistency, scalability, and efficient collaboration with developers, I built the interface using a structured design system approach based on Shadcn UI components and Lucide iconography.
The system was designed to support a complex, data-heavy SaaS environment while remaining flexible and easy to maintain as the product evolves.
Wireframes & Prototyping
After validating the workflows, I designed low-fidelity wireframes to explore layout structures and interaction patterns.
