Fintech · Internal Tool · Anti-fraud · Pix
Ewally
Pix anti-fraud operations
Designing a decision-support workflow for reviewing suspicious Pix transactions, handling refunds, and keeping every case traceable.
Context
Pix made transfers faster and fraud response more critical.
Pix transformed money transfers in Brazil by making them immediate and easy to use. That same speed increased the importance of fraud prevention and operational response.
When a suspicious transaction was reported or flagged, Ewally’s anti-fraud team needed to review the case, understand what happened, and decide how to proceed
The challenge
Turn complex fraud cases into clear operational decisions.
Operators were dealing with sensitive financial information, multiple parties, transaction details, and actions that could lead to approval, rejection, or refund handling.
The interface needed to make the situation understandable at a glance while preserving enough context for a confident decision.
Design focus
✓
Data visibility
✓
Decision support
✓
Traceability
✓
Clear case status
✓
Operational efficiency
My role
UX/UI for anti-fraud operations.
I worked on the internal experience used by Ewally operators to receive, analyze, and resolve Pix-related fraud cases.
My work focused on structuring the workflow, improving information hierarchy, clarifying case states, and designing the interfaces for infringements, decisions, refunds, and historical review.
Workflow
From suspicious transaction to resolution.
01
Infringement received
A suspicious Pix case enters the anti-fraud queue.
02
Case review
The operator analyzes transaction data, payer and receiver information, amounts, identifiers, and supporting details
03
Decision
The case can be approved or rejected, with an operator comment documenting the decision.
04
Refund
handlinglity
When applicable, the case continues into a refund flow.
05
History and traceability
Improve consistency, reduce visual friction and raise the UI standard across the ecosystem.
Case queue
A clear view of incoming fraud cases.
The main list gave operators a structured overview of suspicious Pix cases, including key identifiers, payer information, transaction value, infringement type, and opening date.
The goal was simple: help operators scan and prioritize cases without opening each one individually.
Case analysis
Decision-making with the right context.
The expanded view surfaced the information needed to evaluate a case:
• fraudulent amount
• payer information
• receiver information
• account status
• case details
• notification and transaction IDs
• observations
Instead of presenting raw data, the interface grouped information around the operator’s decision.
Approve or reject
Clear decisions with accountability.
Operators could approve or reject a case and leave comments documenting the reasoning behind the action.
The interaction was intentionally explicit, reducing ambiguity in a workflow where every decision could have financial consequences.
Design principles
04
Clarity under pressure
Sensitive financial operations require calm, readable interfaces with strong information hierarchy.
03
Traceability
Every action needs a clear status, documented outcome, and historical record.
02
Decision support
The interface should help operators interpret a case, not simply expose data.
01
Transparency
Critical transaction and account information should be visible without forcing the operator to search across multiple views.
Outcome
A clearer workflow for sensitive financial decisions
The redesigned experience gave anti-fraud operators better visibility into each case, a more structured decision process, and a clearer connection between infringement review, refund handling, and historical follow-up.
Better case visibility
More structured operator decisions
Clearer status and action history
Stronger operational traceability
Reflection
Fraud operations are not just about data. They are about confidence.
Designing for anti-fraud workflows reinforced an important principle: in high-stakes financial operations, speed matters, but clarity matters more.
The interface needs to help people understand what happened, what they need to decide, and what will happen next
Tom Oliva
Product Designer UX/UI · Buenos Aires, Argentina