Fintech 2024

Treasury Management Dashboard

CFOs were making $2M+ errors because the transaction view didn't surface discrepancies until after approval.

Week 1
Embedded research
Week 4
Synthesis
Week 7
Anomaly-first IA
Week 11
Detail design
Week 15
A/B test
Ship
−62% errors
Treasury Management Dashboard final design
Client
Vaultline Financial
Role
Senior Product Designer
Duration
18 weeks
Tools
Figma, Amplitude, Dovetail, Linear
-62%
Transaction errors
8 min → 3 min
Approval time
+89%
User adoption

The Problem

Vaultline's treasury management platform processed $4B+ in daily transactions, but CFOs and treasurers were making costly errors because discrepancies weren't surfaced until after approval. The transaction detail view buried critical information in dense tables.

  • • Average transaction error rate was 3.2%, costing clients $2M+ annually
  • • Approval workflows took 8 minutes on average
  • • Only 34% of treasury teams used the platform daily
  • • CFOs reported "decision fatigue" from information overload

The Process

I embedded with 3 treasury teams for 2 weeks, observing how they reviewed transactions and where they hesitated. The pattern: they needed to see anomalies instantly, not hunt for them in rows of data.

Discovery

Embedded research with treasury teams

Shadowed 3 treasury teams across 2 weeks. Mapped their decision-making process and identified the "scan and confirm" pattern — they needed to validate transactions in under 10 seconds.

Synthesis

Anomaly-first information architecture

Restructured the transaction view to surface discrepancies first: unusual amounts, new vendors, duplicate payments, and threshold breaches were promoted to the top with visual indicators.

Design

Progressive disclosure for detail

Designed a card-based transaction view that showed summary data upfront, then expanded to show full details, history, and approval context only when needed.

Validation

A/B testing with real users

Ran a 4-week A/B test with 200 treasury users. The new design reduced errors by 62% and cut approval time from 8 minutes to 3 minutes.

The Solution

We redesigned the transaction view as an anomaly-first interface. Discrepancies were surfaced with amber highlights, unusual patterns were flagged with visual indicators, and approval workflows were simplified to a single glance-and-confirm action.

  • ✓ Transaction errors decreased by 62%
  • ✓ Average approval time reduced from 8 minutes to 3 minutes
  • ✓ Daily active users increased by 89%
  • ✓ CFO satisfaction score increased from 2.8/5 to 4.4/5
  • ✓ The anomaly-first pattern was adopted across Vaultline's entire product suite