LAVI — Reflections & Next Steps

LAVI · UX Case Study · Duration: 8 weeks · Role: Certified UX Designer (solo) · Figma VS Code · Section 12 of 12

12: Reflections & next steps

What I learned. Where LAVI goes next.

This case study began with a memory and became a rigorous UX process. Eight weeks of research, design, testing, and iteration taught me things I couldn't have learned any other way. Here's an honest account of what worked, what didn't, and what comes next.

Personal reflection

"The hardest part of this project wasn't the wireframes or the prototypes, it was learning to separate my lived experience from my users' experience. I came in with strong assumptions about what the app needed, because I had watched my parents navigate these exact barriers. Research kept humbling me. The WhatsApp share button? Research surfaced it. I had almost deprioritized it. The body diagram for symptom input? I considered cutting it for simplicity. Marie-Claire's persona told me she would never type a medical word she didn't know. The best design decisions in this project came from listening, not from knowing. That's the lesson I'm taking into every project that follows."

— Rose, Certified UX Designer & daughter of immigrants

Key lessons from this project

Language is the foundation, not a feature

I initially planned multilingual support as a layer added on top of the design. Research made clear it had to be the first decision, shaping every component, every copy choice, and every navigation label from the ground up. You cannot retrofit language access into a design that wasn't built for it.

Emotional state is part of the design brief

Usability testing revealed that the red urgency state was causing anxiety before users had read a single word. That moment reframed how I think about design: the emotional experience of an interface is not a detail, it is the primary experience. Tone, color, and sequence are functional decisions with measurable consequences.

Testing exposes what assumptions protect

The language switcher failing in testing was the most valuable finding in the entire project. I knew where it was, so I assumed everyone would find it. Testing dismantled that assumption completely. What is obvious to the designer is often invisible to the user, the only way to know the difference is to watch someone use the thing you built.

Proximity to a community is not the same as understanding it

Labeling the insurance filter "Medicaid / Medicare" felt neutral to me. To a participant, it felt stigmatizing. Designing for communities I belong to still required active listening and intellectual humility. Shared identity reduces some barriers, it does not eliminate the need for rigorous research.

What I would do differently

Involve community translators from week one

Haitian Creole copy was reviewed by one native speaker during usability testing. Future iterations would involve a dedicated community translation review panel from the research phase, ensuring cultural nuance, not just grammatical accuracy, is embedded from the beginning.

Recruit lower digital literacy participants for testing

All 5 usability test participants were comfortable smartphone users. LAVI's core audience includes users with significantly lower digital fluency, people who rarely use apps beyond WhatsApp. Future testing would specifically screen for this profile, which would have surfaced different and arguably more important issues.

Build the body diagram as an interactive SVG from the start

The body diagram in the prototype was a static illustration. In a production build, each body region would be a tappable SVG zone that pre-fills the symptom field, eliminating the need for users to type medical words they may not know in any language. This was the right design from the beginning and should not have been deferred.

Co-design with community health workers

Promotoras and promotores, community health workers who serve Haitian and Hispanic populations, would have been invaluable co-designers. They hold institutional knowledge about the specific trust barriers, terminology gaps, and care-seeking patterns that research alone cannot fully surface. This is a non-negotiable for v2.

LAVI v2 roadmap

Phase 1. Validation

Community pilot with 20 users over 4 weeks

Partner with Haitian and Hispanic community health organizations in Atlanta and Miami to run a structured pilot. Collect retention data, task completion rates, and qualitative feedback in participants' primary language. Measure whether users are actually taking medications correctly, not just engaging with the app.

Phase 2. Feature expansion

Interactive SVG body diagram + voice input

Build the body diagram with fully tappable SVG zones that pre-fill the symptom field. Add voice input as a primary alternative to typing, critical for users who struggle with written input in a second language. Explore audio medication reminders narrated in Creole and Spanish by community voice actors.

Phase 3. Provider partnerships

Integrate with FQHC provider networks

Partner with Federally Qualified Health Centers to populate the doctor finder with real-time availability and verified insurance acceptance data. Explore telehealth integration so users can book video appointments with language-matched providers directly from within LAVI, closing the gap between finding a doctor and seeing one.

Phase 4. Language expansion

Add Brazilian Portuguese and Arabic support

Brazilian Portuguese represents one of the fastest-growing immigrant populations in the U.S. and shares significant cultural overlap with the existing design system. Arabic expansion would be a larger architectural investment. Arabic is written right-to-left, requiring a mirrored layout system rather than simple translation, but the population need is significant and worth the scope.

RTL layout note: Arabic support requires a full bidirectional layout architecture, not just translation. This is a separate design and engineering workstream and should be scoped independently.

Phase 5. Impact measurement

Track health outcomes, not just engagement metrics

True success for LAVI is not measured in daily active users, it is measured in whether people take their medications correctly, attend follow-up appointments, and understand their diagnoses. Partner with public health researchers to design a longitudinal outcomes study. Build the measurement framework before the product scales.

Skills demonstrated

Core UX competencies

User research Usability testing Persona development Journey mapping Information architecture Lo / mid / hi-fi prototyping Design systems Iterative design

Specializations

Health tech UX Multilingual design Culturally inclusive UX Mobile-first design Health content strategy

Tools and standards

Figma VS Code WCAG 2.1 AA Competitive analysis MoSCoW prioritization Developer handoff

Project deliverables

The complete LAVI Figma project file, including all three fidelity levels, the design system, component library, and annotated developer handoff frames, is available on request. This case study represents the documented UX process; the Figma file contains the full design artifact.

If LAVI existed at scale, projected community impact

700K+

Haitian Creole speakers in the U.S. currently without a native-language health literacy app

Haitian Health Foundation · U.S. Census estimates

62M

Hispanic Americans who could benefit from culturally competent digital health access in Spanish

Pew Research Center, 2023

$3.4B

Estimated annual preventable healthcare costs linked to language barriers and low health literacy

AHRQ, Joint Commission reports

I built LAVI around one belief

"Understanding your own health is not a privilege. It is something every person deserves, in the language they think in, at the level they can access."

Eight weeks of research, design, and testing turned a personal memory into a product that works for real people. This is what UX design is for, and this is the kind of work I want to keep doing.

LAVI

Bay Lavi · Give Life · Dar Vida