A bank that understood Hong Konger's actual financial goals.

Building a bank from the ground up — inside a regulatory sandbox.

Mox Bank launched in September 2020 as one of Hong Kong's first virtual banks, regulated under the Hong Kong Monetary Authority. I joined early in the design process, working across Onboarding, Save, Spend, Growth, and Identity & Access Management — essentially the full product surface of a bank being built from scratch.

This case study focuses on Goals, the centrepiece of the Save proposition — but the story starts with a hard constraint the whole team had to design around.

The HKMA required that no entirely new financial product could launch with the bank. Every feature had to be derived from existing approved banking services. That constraint, rather than limiting us, pushed us to find meaningful innovation in how we presented familiar instruments — and that turned out to be the right problem to solve.

A bank built around the Hong Kong young professional.

The executive direction was specific: Mox would be Hong Kong's bank for young professionals — a generation that travels, that dreams big, that lives in one of the world's most expensive cities, and that has grown up expecting more from technology than traditional banks were offering.

The proposition wasn't to beat existing banks on interest rates. It was to understand this customer's life — their short-term wants and long-term goals — and build a bank that actively helped them make progress toward both. Small wins matter as much as big ones. A trip to Tokyo matters. A new iPhone matters. These are as real as retirement.

The business logic reinforced this. A savings product that truly served customers' goals would also build the kind of daily engagement that kept them coming back. We weren't just holding deposits — we were designing a habit loop, where setting, progressing toward, and achieving Goals gave users a reason to return to the app with purpose. Automated savings rules were central to this: a customer with an active rule was an inherently more retained one — they had a target to track, a plan in motion, a bank that felt like it was working for them.

They weren't unmotivated. They were stuck.

Before the bank's conception, R/GA conducted quantitative research on Hong Kong saving behaviours. Once I joined, I ran iterative qualitative research every two sprints — designing, moderating, and synthesising user interviews and usability tests with mid- to high-fidelity prototypes.

But how people saved was fragile: a salary account for daily in/out, a second for petty cash, a third tucked away for long-term saving — manual transfers between them, across multiple banks, testing willpower every single time.

And the goals themselves were either impossibly large or dangerously vague:

• Buying a flat in Hong Kong requires HK$7–9M for an average flat in a non-prime district. For most young professionals, that goal is so far out of reach it becomes paralysing — they stop saving for anything.
• Retirement feels abstract. Without an immediate trigger, it's easy to defer.
• Monthly targets existed in people's heads, but weren't being followed — purchase impulse consistently beat the intention to save.

The most common pain points that surfaced across sessions: overspending in certain categories without realising until too late; savings plans at traditional banks that locked you in with punishing exit penalties; low returns that made saving feel pointless; and a total lack of visibility across accounts at different banks.

Five ideas that shaped the design.

Working closely with the Head of Accounts & Deposits and the product team, we distilled the research into five design principles — each one addressing a specific failure mode in how people save.

A four-step flow built around momentum, not obligation.

The goal creation flow was the most carefully considered piece of the experience. The order of steps was deliberate — we wanted users to feel inspired before they felt committed.

A note on loading time — one of those details that only surfaces in production: in our UAT environment, Goals were created in about 2 seconds. In production, that stretched to 5 seconds — on the edge of the threshold where users feel disconnected from their action. Rather than waiting for an infrastructure fix, we designed a loading animation that gave the moment some ceremony. The wait became part of the experience. I'd used the same principle in Onboarding, where up to 30 seconds of background processing was anticipated — the motion designer's animations absorbed it, and users felt like something meaningful was happening, not that they were waiting.

Research-led, iteration-tested, shipped with care.

The project ran on a two-sprint research rhythm. Every two sprints, I designed, ran, and debriefed a round of qualitative sessions — user interviews paired with usability tests on mid- to high-fidelity prototypes. Each round informed the next design iteration, which went straight back into testing.

Early design explorations ranged widely. We tested concepts around visual savings "buckets," calendar-based goal timelines, social saving with friends, and gamified progress mechanics. Some of these sparked strong reactions — in both directions. The sessions were less about validating a direction and more about understanding the emotional texture of saving: what made it feel meaningful, what made it feel like a chore.

That shift — from the destination to the starting point — led directly to the inspiration-first flow. By showing achievable, joyful examples before asking for any commitment, we lowered the activation energy for creating a first Goal. Later rounds of testing confirmed that the order of steps mattered as much as the steps themselves.

Throughout, I worked closely with the visual designer on the UI detail — the avatar personalisation, the progress treatment — and the motion designer on how animations could mark moments of progress and achievement without feeling gratuitous. The UX writer shaped the tone on the inspiration screen, which had to feel encouraging without being patronising. Each of those collaborations shaped the outcome as much as the structural design decisions did.

Stakeholder alignment ran in parallel. I facilitated an Inception Workshop with product owners, engineers, and the Head of Deposits to share research findings and map a shared understanding of the user journey. Getting everyone in the same room, reading from the same research, made the subsequent design reviews go faster and the eventual delivery go smoother.

Goals shipped — and gave us a feedback loop worth following.

Goals launched as one of Mox's signature Save features. Before launch, the whole team aligned on a clear set of success metrics to measure whether the design was working — not just whether it was used.

Post-launch, the data opened new questions worth designing for: when do customers create their first Goal after onboarding — and are there seasonal patterns? What do people name their Goals? Are target dates being hit? How does automation uptake correlate with goal completion over time?

Those questions fed the next iterations — improving the inspiration examples to lift adoption from 13% toward 20%+, designing milestone celebrations, and building re-engagement nudges for Goals that had stalled. The product didn't stop at launch; it became a system we could learn from and improve.

What I carry from this.

Working at Mox was where I learned what it means to build a product people trust from scratch. When a bank is new, every onboarding screen, every balance view, every error state is part of the argument for why someone should trust you with their money. There's no legacy to coast on. The design has to earn it.

Goals specifically taught me that the best user insights often reframe the brief entirely. The shift that mattered most — from designing for the goal to designing for the first step — only emerged from being in the room with users, watching where confidence broke down. No amount of desk research would have surfaced it.

And I came away with a genuine appreciation for the team around me. The motion designer's animations made progress feel alive in ways a static prototype couldn't describe. The UX writer's copy on the inspiration screen was what made users feel seen. Our engineers found creative solutions within real infrastructure constraints. Good design at this scale is genuinely collective — my job was to hold the vision and make the space for everyone to do their best work within it.