Vaulta Wallet History

Instead of starting with UI improvements, I reframed the problem from the admin’s perspective. Admins don’t manage a single merchant. They manage many similar to how a hotel headquarters oversees multiple branches under one roof.

Each merchant behaves differently: some grow steadily, some fluctuate within a normal range, and some show sudden spikes or drops. In this context, admins don’t have time to read tables or compare values row by row. They need to scan for movement.

This shift in perspective became the foundation of the redesign.

At this stage, I intentionally did not move into UI exploration.The core issue wasn’t layout, typography, or table density. It was that the system itself placed too much responsibility on the admin.

In a product operating at scale, interpretation should not be manual.
The system should:

• Detect meaningful movement
• Seperate noise from signal
• Surface deviations
• Guide attention
• Reduce congnitive effort

So before designing screens, I focused on “defining how the system should think.” This reframed the challenge from UI design to financial intelligence design.

Previously, interpretation lived entirely with the admin.

Now, This redesign moves interpretation into the system, reducing cognitive load and enabling faster, more confident decisions.

Rather than exposing raw transactions and balances, I designed an interpretation layer that sits between data and interface.

This layer is responsible for:

• aggregating transactions
• calculating movements
• normalizing change
• classifying severity
• communicating risk consistently

The UI becomes the “output of this system”, not the place where interpretation happens. This shift turns the dashboard from a “reporting tool” into a “decision-support system”.

To represent movement clearly, I introduced a new primary signal “Net Change”.

It compresses wallet activity over a defined time period into a single,
scannable signal by combining:

• Absolute value (₹) - the total amount of money that moved
• Percentage (%) - normalized movement relative to the opening balance
• Direction (↑ / ↓) - whether the balance increased or decreased

By anchoring change to a specific time window and normalizing it across merchants, Net Change allows admins to immediately understand what changed without manual comparison or mental calculation.

A critical design decision was to decouple direction from severity.

1. Direction answers: *What happened ?*

2. Magnitude answers: *How serious is it ?*

In financial systems:

• a large positive spike can indicate fraud or reconciliation errors
• a large negative drop can indicate refund spikes or merchant distress

Both are operational risks.

These thresholds are designed to absorb expected financial volatility, reduce false positives, surface meaningful anomalies, and maintain trust in system decisions, with arrows indicating direction and severity conveying risk.

Detecting movement alone is not enough.Admins also need to understand why the system flagged a merchant.

To support this, I designed an Explanation Layer that translates system analysis into simple reasoning. When an admin opens a merchant row, the system answers three quick questions.

1. Primary Signal: Net Change

Net Change is the primary signal surfaced by the system, combining value, percentage, and direction into a single, scannable unit.

This final dashboard brings together movement detection, risk interpretation, and prioritization. Admins act on decision-ready intelligence rather than raw data.

“How does this intelligence fit into the broader Nexdha system?”

The interpretation layer sits between Vaulta transaction engine and the admin interface.

It processes wallet activity once, centrally, and outputs consistent intelligence across tables, filters, and summary metrics. This ensures that every admin view is driven by the same logic, reducing inconsistency and manual interpretation.

Before

• 1. Admins manually compared balances
• 2. Risk assessment relied on intuition
• 3. Monitoring required time and effort

After

• 1. The system surfaces abnormal movement automatically
• 2. Risk is Classified Consistently
• 3. Admin focus on Investigation & Action
• 4. Introduced a reusable interpretation layer

This project was not about redesigning a table.

It was about deciding what a financial system should take responsibility for, and what it should no longer ask the user to interpret. By introducing movement, context, and explainability into the system itself, the Wallet History evolves from raw data into a source of intelligence.

“This is where product design transitions from arranging interfaces to designing systems of decision-making.”