Health & Wellness App

Transforming Retention Through Experience Design

How do you turn a broken flagship feature into a retention engine? I reframed “fix the timer” into a scalable dashboard vision—shipping a unified sober timer and widgetized home that rebuilt trust, simplified onboarding, and created a platform for future features.
Week‑4 retention 5%→22% Cross‑platform consistency AA by default Support tickets −60%

Context

This Health & Wellness App is a peer-support app for people in recovery. Over four years, it reached 1M downloads and 150K daily users, but retention collapsed: only 5% of new users were active four weeks post-download, despite increased ad spend.

My Role

As Lead UX/UI Designer, I was tasked with reversing retention decline, aligning product expectations with delivery, and creating a design system scalable to future features.

The Challenge

User research surfaced three systemic problems:

  • Onboarding was onerous : 10–15 minutes, causing drop-off.
  • Expectation mismatch : advertised features didn’t deliver, leading to fast abandonment.
  • Fragmented design : Android and iOS experiences diverged, creating confusion and eroding trust.

Stakeholders wanted results fast, but the risk was alienating the loyal core audience.

Strategic Decision

Rather than tackling everything at once, I focused on the sober timer, the app’s flagship feature. Research showed many users downloaded the app primarily for a reliable sobriety tracker. If we failed them here, no downstream feature mattered.

Sober timer existing Android
Initial (Android) timer layout looks broken and suffers from accessibility issues.
Sober timer existing iOS
Initial (iOS) timer layout is confusing and messy.

Not only was the sober timer different on each platform, it was a complete mess.

Process

Competitive Benchmarking

  • Conducted a structured audit of 12 competing sobriety apps.
  • Built a feature matrix highlighting strengths (e.g., milestone celebrations, personalization) and weaknesses (e.g., cluttered dashboards).
  • Synthesized findings into design principles: clarity, consistency, celebration.

User Insights

  • Interviewed 15 users who had churned within 30 days.
  • Findings: the timer felt unreliable, inconsistent across platforms, and lacked emotional resonance.
  • Quote: “If the timer doesn’t work right, why should I trust the community?”

Design & Iteration

  • Wireframes tested with 20 users to validate information hierarchy.
  • Accessibility audit led to improved contrast, tap targets, and VoiceOver compliance.
  • Iterated toward a modular widget framework, turning the timer into the first in a suite of configurable dashboard widgets.

Stakeholder Alignment

  • In a key leadership meeting, I pitched: “What if the timer wasn’t just fixed, but became the foundation of a scalable dashboard ecosystem?”
  • This shifted the project from a patch to a platform vision, gaining buy-in for a multi-phase rollout.

The Dashboard

The core of the solution was the dashboard; a widgetized framework allowing users to rearrange and personalize their experience. This created a scalable design system that would enable future features (gratitude journals, streak tracking, peer-support, and more).

Dashboard wireframe
Documentation calling out how the dashboard would adjust for different screens and how a user would edit the position of widgets.
Dashboard wireframe
The sober timer widget in a variety of configurations (style and size).
Dashboard wireframe
Using the dashboard and widgets on several screens to maximize consistency.

Outcomes and Impact

Dashboard wireframe
The number of active users accelerated after V3.0, when we introduced the dashboard.
  • Retention quadrupled: from 5% → 22% four-week active users post-launch.
  • 250% increase in returning users directly attributable to timer redesign (validated via cohort analysis).
  • Established a widget framework that became the backbone of all future feature launches.

Reflection

  • What worked: Anchoring design decisions in user research, tying deliverables to business metrics, and reframing a tactical fix into a strategic platform shift.
  • What I’d do differently: Formalize cross-functional rituals earlier. Engineering initially struggled to adapt to design-driven iteration, causing churn I could have preempted.

Takeaway

Design is not decoration. By aligning user needs (trust in the timer) with business goals (retention + ad ROI), we delivered both immediate wins and a scalable framework for growth.