BitoDebt Β· Fintech Β· 2019β2020
From experiment to 25% of company revenue
I designed Taiwan's first fixed-income digital asset product β from MVP to market leader, through three iterations of trust, compliance, and product-market fit.
ROLE
Design Owner β Research, Strategy, UI/UX, Visual, Brand
TEAM
2β6 people Β· Legal, PM, Eng, CS
DURATION
1 year
+34%
CONVERSION LIFT
100%
SELL-OUT RATE
+57%
TRADING VOL.
25%
COMPANY REVENUE
Challenge
Nobody trusts a product they've never seen.
Taiwan's first fixed-income digital asset product: users deposit stablecoins, receive principal plus fixed interest (8.5% APY) at maturity.
The problem wasn't usability β it was courage. Users understood the flow, but didn't dare press "Subscribe."
βAfter it matures, how do I get my money back?β
β Participant P8
βLocked for too long β what if the exchange goes down?β
β Participant P9
βI'd feel more confident once more people in Taiwan are talking about it.β
β Participant P7
Iteration 01
Ship first, don't chase perfection.
Judgment: Strip it down to the skeleton.
Team of 2, strict legal compliance. I stripped the product to minimum: one subscription flow, plain language, zero visual packaging.
The hardest part wasn't designing the flow β it was naming the product. "Debt rights" was the legally approved term β accurate, compliant, but users couldn't understand it. I spent weeks negotiating with legal: they guarded compliance, I guarded user comprehension. Final decision: keep the legal name, but translate it into plain language at every touchpoint.
MVP wireframe β one flow, nothing extra.
Result
Launched successfully. Conversion rate: 22%.
Insight
Users came and looked, but didn't dare press "Subscribe."
Iteration 02
Make the abstract tangible.
Judgment: Trust matters more than flow.
Data showed users didn't distrust the flow β they distrusted the concept. I needed to make "fixed-income digital assets" feel real, not scary.
The Magic Bean concept wasn't in anyone's brief. I saw the product needed a visual story to lower the cognitive barrier, created a Mood Board on my own, then handed the entire visual system to the marketing team to extend. That's design ownership in a team of 2.
Result
Page views Γ2.5β3.5, longer dwell time, more users entering subscription flow.
Insight
But conversion didn't scale proportionally β more visitors, but completion didn't keep up.
Iteration 03
My assumption was hurting users.
Judgment: Admit, fix, learn β same day.
After launch, customer service reported an emergency: "Users are furious β they can't find the BITO payment option." I had assumed most users wouldn't hold much BITO (platform token), so I defaulted to USDT. That assumption was wrong. 75% of long-term users held significant BITO and wanted the fee discount (20% β 10%). Some had already paid the higher fee. In fintech, a wrong default isn't "inconvenient" β it causes real losses.
Just swapped left and right, changed the default. Fixed same day. Support complaints dropped to zero.
βThanks, I finally bought it.β
β Telegram
βThis is exactly what I wanted β simple and straightforward.β
β Telegram
Key Finding
Trust can't be designed β only proven.
After fixing the payment issue, I conducted a second round of user interviews. The finding changed everything: trust didn't come from my design. It came from three external signals.
Strategy Shift
Timeline
Phase 1
MVP β Ship First
Users didn't understand the concept. We chose a minimal MVP β kept only the core purchase flow, removed all non-essential info, let users complete the basic operation first.
βLet the product survive first, then educate the market.β
How I Collaborate
No authority, still drove change.
"Naming alone was a battle"
"Debt rights" was what legal required β accurate, compliant, but users couldn't understand it. Weeks of back and forth. They held the line: compliance. I held the line: user comprehension. Final call: keep the legal name, translate at every touchpoint.
"I initiated the brand, they amplified it"
The Magic Bean visual system wasn't marketing's brief. I created the Mood Board, built the concept, then handed the visual language to marketing to extend across campaigns.
"My fastest user research channel"
After the BITO payment incident, I joined user Telegram groups, attended weekly CS meetings, recruited interview participants through CS. The most critical design fix came from a forwarded complaint β not from data analysis.
Results
From experiment to core business.
CONVERSION LIFT
22% β 56%
SELL-OUT RATE
Sold out within a day
TRADING VOL.
Exceeded target
COMPANY REVENUE
Of company total
Years later, the live version is still my original design.
βSince using BitoDebt, I can finally sleep in peace. The anxiety of waking up at midnight to find my assets cut in half is gone. Now 30% of my crypto portfolio is in BitoDebt β it gives me stable returns and peace of mind.β
β Bob, long-term user
Reflections
Four things I learned building a trust product.
Challenge your most confident assumption.
The payment default error taught me: in fintech, a designer's intuition is the most dangerous thing.
Trust can't be designed β only proven.
No visual metaphor beats a real user's feedback screenshot. What a designer can do is not block the sources of trust.
The fastest research channel is inside your company.
The most critical design fix didn't come from interviews or analytics β it came from a forwarded complaint from CS. But that wasn't luck β I had proactively built that relationship.
Design ownership means seeing the gap and filling it.
The Magic Bean concept wasn't in anyone's brief. I saw the need, did it myself, then handed it to marketing to scale.