Project Overview
The Customer Due Diligence (CDD) application is an internal tool designed to support compliance agents in verifying and processing sensitive customer data as part of Evoke’s regulatory and fraud prevention protocols.
My responsibility was to lead the UX and UI efforts for a platform that transforms fragmented workflows into structured, task-based processes. The system centralizes decision-critical information—previously spread across multiple channels—and ensures that every case follows a consistent and auditable path.
Role: Lead UX/UI Designer
Tools: Figma, OutSystems, Miro
Duration: 6 months (end-to-end delivery)
Objectives
- Design a scalable, role-based interface for internal due diligence
- Replace inconsistent and manual workflows with guided, compliant task flows
- Improve visibility into task ownership, history, and performance
- Standardize data input while reducing cognitive load on agents
- Enable future integration of AI and automation layers
Discovery & Research
To ground the product in real user needs and business constraints, I collaborated directly with stakeholders across compliance, risk, and operations.
Key activities:
- Process shadowing with agents to understand real-world bottlenecks
- Stakeholder interviews with Legal, PMO, and IT for flow mapping
- Evaluation of existing spreadsheets, disconnected tools, and audit processes
- Persona creation for agents, reviewers, and administrative users
Pain points identified:
- Case status was manually tracked and often unclear
- Data lived in separate systems, slowing down reviews
- No clear ownership or logs for decision-making
- High potential for human error under pressure
UX Strategy
To address these issues, I proposed a modular, task-based experience that guides agents step-by-step through the full due diligence flow.
Core flows included:
- False Positive Check
- Customer Information Validation
- Address Confirmation
- Summary, Risk Insights & Final Recommendation
Each task was intentionally scoped to reduce overload and clarify what action was needed at every stage. I introduced:
- Status indicators per task (assigned, in progress, completed)
- Role-based actions (claim, release, resume, complete)
- Contextual UI elements based on task type and progress
- Task logs for compliance traceability
UI Design
The system was designed to handle complexity without overwhelming the user. I created a design system in Figma, tailored to enterprise-grade operational tools and capable of scaling for future features.
Highlights:
- Over 30 unique screens designed across workflows, modals, dashboards, and exports
- Reusable components for form fields, validations, navigation, and notifications
- Interfaces optimized for keyboard use, quick scanning, and clear error feedback
- Information prioritization for decision-critical fields (e.g. MOH Score, risk triggers)
- Style guide compliant with accessibility standards (WCAG)
The UI structure is adaptable to both desktop workstations and tablet-based interfaces, supporting a hybrid workforce model.
Usability Testing
I conducted moderated usability tests with global users from LATAM and EMEA.
Key findings:
- Users struggled with long, scroll-heavy layouts → changed to tabbed navigation
- Agents were unsure about the implications of actions → added tooltips and inline help
- Difficulty tracking task progress → introduced persistent visual timeline indicators
These iterations led to significantly better engagement and reduced support requests.
Implementation & Handoff
Beyond design, I collaborated actively with developers throughout the build process:
- Delivered fully annotated Figma files with interaction logic
- Validated frontend components during sprint reviews
- Supported integration of Single Sign-On (SSO) and role-based access logic
- Participated in QA to ensure logic alignment and error state consistency
This close collaboration ensured a smooth rollout and allowed for rapid post-launch improvements.
Outcomes
- Reduced average case handling time by 30%
- Increased agent confidence and ownership through clearer workflows
- Greatly reduced error rate and manual escalation
- Compliance team reported faster and easier audit reviews due to traceable task logs
- System structure is now being reused for additional processes (KYC updates, AML alerts)
Ongoing Development
The CDD application is now entering its second phase. We are currently designing and testing AI-powered features that will:
- Autofill known customer information based on past records
- Suggest next best actions based on behavior patterns and risk factors
- Surface inconsistencies and flags proactively
- Support real-time guidance without removing control from human agents
This layer of automation is being designed with transparency and auditability in mind, preserving agent accountability while improving efficiency.