Failure 01
Return-only ranking misclassifies risk-takers as skill
A 312% return over three weeks of leveraged Bitcoin reads identical to a 312% return over four years of disciplined trading. Followers can't tell which one is repeatable.
ACY · COPY-TRADING PLATFORM
TradingCup
Designing trust for retail copy-trading
A copy-trading platform where 10,000+ retail followers select algorithmic strategy providers. Sole product designer for the entire surface — provider evaluation cards, pre-allocation disclosure flow, follower risk controls, leaderboard staging, brand identity. The thesis: the first loss is the design problem, not the sign-up screen.
Where this sits — Scale 02 of 05
Before Xanthos. After ACY. ACY proved compliant infrastructure can hold across 40+ regulated markets. TradingCup proved compliance is necessary but not sufficient — you also have to design for the psychology of a follower under loss. The insight that emotional state during drawdown is a design variable carries directly into the Xanthos Private Bank thesis, where the first bad quarter is the precise moment that makes or breaks a $28M advisory relationship.
The Problem
Three failures retail copy-trading platforms inherit by default
When a platform offers to "follow successful traders," the design either prepares the follower for what will actually happen, or it doesn't. Most don't. Each failure below is something I watched users do at session-recording level.
Failure 02
First drawdown breaks the relationship before it starts
In session recordings, most followers who churned did so on their first sustained loss — not because the loss was large, but because nothing in the sign-up flow prepared them for normal volatility.
Failure 03
"Recommended" is a regulated word, not a UX word
The first cut of "Recommended Providers" was killed by Legal at pixel-perfect stage — implied endorsement under ASIC inducement rules. Pivoted to "Risk-Adjusted Ranking" with disclaimers; shipped clean.
Design Thesis
The first loss is the design problem
Most copy-trading platforms optimise the sign-up funnel. The actual relationship dies at the first sustained drawdown — not at registration. So I designed the entire surface backwards from that moment: every screen prior has to prepare the follower for it.
Most copy-trading platforms treat onboarding and resilience as two products. They are one product. The follower who signs up but quits at week 11 was acquired and then lost — and the design owned both.
Internal design memo · TradingCup v2 framing
Before
Optimise sign-up. Hide drawdown until after deposit.
After
Pre-allocation disclosure. Drawdown shown before any capital decision. Trust hierarchy on every provider card.
Provider Evaluation Framework
Four metrics, ranked by what protects the follower
The provider card surfaces the same four metrics in the same order on every surface: list, detail, search, comparison. The order is not alphabetical — it's risk-first. Return is fourth, not first.
Metric 01
Maximum drawdown
Peak-to-trough percentage loss. Surfaces first because it answers the only question that matters before deposit: how bad does this strategy get on its worst day?
Metric 02
Sharpe ratio
Excess return per unit of volatility. Tooltips translate it as "extra reward earned per dollar of risk" — the explanation has to land in one read for non-CFA followers.
Metric 03
Win rate & consistency
Percentage of profitable trades. Paired with average win/loss size to prevent the "high win-rate but one massive loss" decoy. The number itself is intuitive; the pairing is what makes it honest.
Metric 04
Time on platform
Months of verified trading history. A 38-month track record survives at least one volatility regime; a 6-month one might just be lucky. Verification badge unlocks at 24 months minimum.
Pre-Allocation Flow
The disclosure screen before any capital decision
Three screens between provider selection and deposit. Each one is an off-ramp, not an upsell. Drop-off here is a feature — followers who quit at the disclosure screen are followers who would have churned at week 11 anyway.
Step 01
Historical drawdown disclosure
A static chart of this provider's worst 6 drawdowns over the lookback window, with duration and recovery time labelled. Followers who saw this screen retained at higher rates than the cohort that didn't.
Step 02
Acknowledgement (not a checkbox)
A free-text field: "What's the largest loss I'm prepared to accept on this allocation?" The follower types a number. The system uses it later to set their auto-pause threshold.
Step 03
Conservative defaults, every time
Allocation cap defaults to 5% of account equity. Copy ratio defaults to 1:1. Both are explicit choices the follower has to override — not the most aggressive default with a "lower it if you want" hint.
In-Position Controls
Three follower-side controls the provider cannot override
Once a follower is copying, the design has to honour the principle that the follower's risk budget is the follower's, not the provider's. Three controls that the lead trader cannot turn off.
Control 01
Per-trade stop-loss
Follower-set hard stop on every copied trade. Independent of the provider's stop. Defaults conservative; can be tightened, never loosened beyond the platform max.
Control 02
Allocation cap
Maximum percentage of account equity any single provider can use. Default 5%. Hard ceiling enforced by the platform — followers cannot override beyond 25% even with confirmation.
Control 03
Auto-pause on threshold
When cumulative loss on this provider crosses the follower's stated tolerance (from the pre-allocation form), the system pauses copying automatically. Resume requires an explicit human action.
Compliance Pivot
When Legal kills your pixel-perfect feature, the redesign teaches you something
"Recommended Providers" was scrapped at handoff for ASIC inducement risk. The replacement — "Risk-Adjusted Ranking" — survived a two-year audit cycle with zero design-related findings. The pivot itself is the case study.
Before · killed by Legal
"Recommended Providers"
- Implied platform endorsement
- ASIC Corporations Act s.911A inducement risk
- 4 weeks of high-fidelity work scrapped
- Brand exposure if shipped
After · approved & shipped
"Risk-Adjusted Ranking"
- Algorithm-derived, transparent
- Methodology disclosed in tooltip + footer
- "Past performance does not predict future results" pinned to every card
- Two-year audit window · zero findings tied to the surface
Process change after this pivot: Legal joins the Figma file at wireframe stage on every regulated feature. Three hours of friction at the start prevents three weeks of rework at the end. This was the seed of the "Legal-first design" methodology that carried into ACY's compliance system.
Outcomes
What the surface produced over the two years it ran
Numbers below are platform-attributable, not design-attributable in isolation. Design contribution = the surface that converted, retained, and stayed compliant.
187K+
Organic search clicks · 12-month period
Source · Google Search Console
42.37%
GA4 engagement rate · 30-day rolling
Source · Google Analytics 4
10K+
Active followers on platform
Platform-stated · design = scalability
0
Design-related ASIC findings · 2 yr
Internal regulatory audit
Cross-Functional
How the work was coordinated
Sole product designer for the entire surface. The work was not solo — it was orchestrated across PM, engineering (Sydney + Taipei), Legal, Compliance, and the trading desk.
Engineering
Sydney + Taipei async pipeline
Bilingual specs (English UI labels + Chinese rationale notes) eliminated the translation tax. Loom walkthroughs replaced the synchronous handoff meetings that 4-timezone teams can't have.
Legal
Joint Figma file from wireframe stage
Legal counsel reviewed clickable prototypes — not static mockups — at every regulated feature. Disclaimer wording, tooltip copy, and label semantics were jointly drafted in the same file as the design.
Trading Desk
Provider taxonomy validated by quants
The four-metric framework (Max DD, Sharpe, Win rate, Time on platform) was cross-checked with ACY's prop trading desk. Quants vetoed two earlier drafts that buried drawdown behind return.
Constraints & Reflection
What I had to compromise on, and what I'd do differently
The pre-allocation flow reduced sign-up conversion. That was deliberate. Other constraints were less defensible — listing them is part of the work.
Constraint 01
Sign-up conversion fell — and that's the trade
The pre-allocation disclosure screen reduced funnel conversion. Retention at week 12 went up. The PM and I agreed the trade was worth it; the CFO needed convincing with cohort data.
Constraint 02
Mobile-first compromised by chart density
The provider evaluation card needs four metrics visible. On mobile the layout collapsed to a tap-to-expand pattern. Ideally the four metrics would be glanceable on every viewport — they're not.
Reflection
What I'd do differently
More follower-side qualitative research before the v1 ship. The pre-allocation flow was correct in retrospect, but I designed it from session recordings and intuition, not interviews. Senior PD work earns the next role by closing that gap.
What this case study reads as for a senior copy-trading PD role
Sole-designer ownership of a regulated retail-finance surface that scaled to 10K+ followers under ASIC oversight, with a coherent thesis (the first loss is the design problem) that produced shippable artefacts (provider card hierarchy, pre-allocation disclosure flow, follower-side risk controls), survived two years of regulatory audit, and integrated across PM, Legal, engineering, and the trading desk. The interactive demo below is one of those artefacts.
Live Demo · Provider Evaluation System
The counterintuitive finding: showing risk data increased sign-ups
The initial brief was to hide drawdown data. Research said otherwise: users who saw the full risk profile signed up at 31% higher rates than those who saw return data only. The design decision was to build a trust hierarchy — surface the most important risk signal first, then let users drill down. This is that system.
Click any provider card to expand · Sort by metric · Scores refresh every 4s · All data simulated
Sort by:
Risk tolerance:
Portfolio threads
Where this case study sits in the larger web
Every problem we solve for clients has multiple valid approaches — different costs, different ROI, different risk profiles. These threads show how the approach on this page compares to others in the portfolio.
Retail → Institutional Translation
Taking consumer-grade UX expectations into regulated/professional contexts — or reverse-porting institutional discipline back to retail.
- Finlogix · Retail Trader Education Retail trader education Low eng cost · A/B tested · Cohen's d 2.47
- ACY Connect · B2B Compliance Bridge Retail broker → B2B platform High eng cost · 12+ institutional clients
- ACY Securities · Regulated Broker System Retail-facing regulated broker High eng cost · 100K+ traders · 40+ jurisdictions
- Trading Cup · Competition Staging Competition gamification Low eng cost · Retail engagement · live leaderboard
- Logix Panel · Broker Operator Console Operator-side of retail platform High information cost · Ops control density
Editorial Voice in Finance
Luxury, editorial, and brand discipline applied to financial interfaces — where restraint itself is signal.
- Christie's · Private Client 260 years of earned restraint High brand cost · Luxury typographic voice
- Nova · Institutional Trading Terminal Terminal chrome as brand Mid-tone palette · mono typography · zero decoration
- Logix Panel · Broker Operator Console Maximum information density per pixel High information ROI · density over whitespace
- Private Banking Advisory Advisor disclosure letters Quiet room · wealthy audience · editorial cadence
- Trading Cup · Competition Staging Event-grade polish High staging cost · Short-lived · High-traffic moment