Loan management
on a blockchain
Trade finance still runs on email, fax, and phone calls. TradeiX set out to put buyers, suppliers, and banks on one automated ledger, and I designed the loan management system that made it usable.
Finance still runs on fax
The Marco Polo Network moves trade finance transactions with security built into the ledger itself. My job was the loan lifecycle it needed: setup, approval, drawdown, and settlement, designed so a banker could trust it on the first pass and a supplier could finish it without training.
Three parties, no shared view. Buyer, supplier, and bank each worked from their own records, status lived in someone's inbox, and every step needed chasing. Result: delays nobody owned.
Manual data, drifting numbers. Figures re-keyed at every handoff meant stock and loan records could disagree, and paper settlement slowed everything down further.
Banking rules are not optional. Every action needs a second pair of eyes, audit trails must hold up to review, and security standards constrain the interface. Compliance had to be designed, not bolted on.
My role & process
One of two designers, working with the CEO, product owner, product manager, and engineering. I owned the loan lifecycle: setup, approval, drawdown, and settlement.
- 01
Discover
Stakeholder interviews and a study of the domain and its compliance rules. I learned how banks approve a loan before designing a screen for one.
- 02
Define
Early alignment with the CEO, product owner, and product manager on scope, then an information architecture built around three parties sharing one loan.
- 03
Design
Mapped the happy and unhappy paths for every lifecycle stage before drawing screens, because unhappy paths are where finance software is actually judged.
- 04
Sign-off
Nine months from concept to stakeholder sign-off, on a shared design system that let two designers move like four.
Goals & research
Automate loan setup so parties chase each other less, make every party's position visible to the others, cut cost by making settlement paperless with eSign, and design maker-checker approval into the flow rather than around it, all without slowing stakeholder sign-off.
Stakeholder interviews, CEO/PO/PM alignment sessions, domain and compliance study, and happy/unhappy path mapping all pointed the same direction: in banking, the second approver is the product, not an edge case, trust comes from showing state rather than reassuring copy, and aligning with the CEO early bought speed later.
Strategy into interface
Three principles carried the design: make the loan's state legible to every party at a glance, build approval into the path so compliance never feels like a detour, and remove paper wherever a signature was the only reason for it.
Key decisions
- 01
Design maker-checker into every stage
Banking requires two people to complete a transaction. The second approver is the product, not an edge case, so every stage shows who submitted, who verified, what changed, and what is still waiting.
- 02
Show state instead of reassuring
Trust comes from showing where things stand, not from reassuring copy. Each party sees exactly where the loan is and what it needs from them next.
- 03
Settle with eSign, not a courier
Paper stayed only where a signature was the reason for it. Documents are generated from templates and signed in place, so settlement closes inside the platform.
Explore the work
Learnings
Regulated products reward designers who learn the rules first. Multi-party software lives or dies on shared state, and mapping unhappy paths early saved months of rework. A shared design system let two designers move like four.
A loan process that ran on email and paper became a single automated lifecycle that banks were willing to put their name against. Designing compliance into the flow rather than around it is what made a blockchain trade finance product feel like ordinary, trustworthy software.