TL;DR
The outcome:
Reduced onboarding from 10+ screens to 3 core steps
Proposed progressive disclosure strategy to minimize drop-off
Navigated technical constraints (MX vs US users, card distribution logic)
What went wrong:
Stakeholders pushed back on delayed data collection due to compliance concerns
Product didn't get traction post-launch (acquisition team issues)
Never got metrics to validate the redesign
What I learned:
How to design defensively when you can't test with users
The importance of documenting trade-offs and UX debt
Stakeholder alignment is as important as the design itself
Main challenge
Context: Loyalty program onboarding asked for everything upfront: personal info, address, documentation, card selection—10+ screens before users saw value.
My hypothesis: Progressive disclosure would reduce drop-off, but I had to design without:
User research budget
Analytics from previous version
Ability to A/B test
The constraints that shaped everything:
Needed address early to route MX vs US users
Documentation upload required for compliance
Stakeholders prioritized data collection over UX
No acquisition team = couldn't validate with real users
Screenshots TBD
Step 1: Light entry, value-first Step 2: Smart routing based on location Step 3: Documentation only when needed
Key decisions:
Compromised on address placement (earlier than ideal, needed for card logic)
Designed "defensively" using heuristics + best practices
Documented trade-offs for future iterations
The pushback:
Stakeholders wanted all data upfront. I advocated for progressive disclosure with compliance examples from similar fintechs. We compromised: kept 3-step structure, moved some fields earlier.
What went wrong?
Product launched with limited distribution. No users = no validation. The redesign never got tested in the real world.
Screenshots TBD
About constraints: Real-world design often means compromising. The skill is knowing WHAT to compromise and documenting WHY.
About advocacy: I should have pushed harder for even 5 user tests. Budget constraints don't mean zero research is acceptable.
About metrics: This experience showed me why product-market fit matters. The best UX design can't save a product without distribution.
For next time:
Even in time/budget constraints, fight for minimum viable testing
Document trade-offs explicitly so future designers understand the debt
Align with stakeholders on success metrics BEFORE designing
Screenshots TBD
