Case Study: Meditate UX — Usability Research & Mobile Redesign

1

Problem Statement

Meditation apps are booming, but user retention is abysmal — industry data shows 77% of users abandon meditation apps within the first week. The app in question had a beautiful visual design but suffered from poor information architecture, confusing onboarding, and a session setup flow that required 7+ taps to start meditating.

As a Certified Usability Analyst (CUA) through HFI Training, I conducted a formal usability evaluation using established heuristics, then led the complete redesign from research through high-fidelity prototyping.

Core Challenge: Reduce friction in the meditation session flow while maintaining the calm, mindful aesthetic — and prove the improvement through rigorous usability testing with measurable metrics.

94%

Task Completion Rate

71%

Fewer Taps to Start

4.2s

Avg. Time to First Session

0

Critical Usability Issues

2

Usability Audit (HFI Methodology)

Using the HFI Certified Usability Analyst methodology, I conducted a structured evaluation combining heuristic inspection, cognitive walkthrough, and competitive analysis. The audit identified 23 usability issues across 5 severity levels.

Heuristic Evaluation Results

HFI Heuristic	Issues Found	Severity	Example
Visibility of System Status	3	Medium	No loading indicator during session initialization
Match Real-World Conventions	4	High	"Binaural" toggle with no explanation of what it does
User Control & Freedom	5	High	No way to pause/resume without losing session progress
Consistency & Standards	4	Medium	Settings accessed via 3 different icon styles across screens
Error Prevention	3	High	Accidental swipe dismisses active meditation with no confirmation
Recognition over Recall	4	Medium	Users must remember duration/frequency settings between sessions

Critical Finding: The session setup flow required 7 sequential taps across 3 different screens. Competitive analysis showed Calm and Headspace achieve the same in 2 taps from the home screen.

3

User Research & Insights

I recruited 12 participants across 3 user segments (beginners, occasional meditators, daily practitioners) for moderated remote testing sessions. Each session included think-aloud protocol, task completion metrics, and post-session System Usability Scale (SUS) questionnaire.

Key Research Findings

83% Confused by Onboarding

"I just want to start meditating, why do I need to answer 10 questions first?" — The onboarding collected unnecessary preference data before allowing any interaction.

No Session Persistence

Users who accidentally closed the app lost their entire session. No auto-save, no resume prompt. This was the #1 cause of frustration in testing.

Duration Guessing Game

Setting meditation duration required scrolling through a wheel picker with no visual preview. Users couldn't tell if they'd selected 5 minutes or 50 minutes.

"I tried to meditate for 5 minutes before work. It took me longer to set up the session than to actually meditate." — Research Participant #7, Beginner Meditator

4

Redesign: Wireframes to High-Fi

The redesign followed a three-phase approach: information architecture restructuring, wireframe validation through rapid testing, then high-fidelity design with micro-interactions.

Phase 1 — IA Restructuring

Collapsed the 7-tap session flow into a "Quick Start" card on the home screen. Moved advanced settings (binaural beats, session length, ambient sounds) into a single "Customize" bottom sheet. Removed mandatory onboarding — users can start immediately with sensible defaults.

Phase 2 — Wireframe Testing

Tested 3 wireframe concepts with 8 participants using Maze for unmoderated task testing. The "Card-Based Quick Start" concept achieved 92% task completion (vs. 67% for the original flow). Validated the single-screen session setup with inline duration picker.

Phase 3 — High-Fidelity Prototyping

Built the final prototype in Figma with 15 connected screens, Lottie micro-animations for breathing guides, and a design system using Montserrat typography, a calming blue-green palette, and WCAG-compliant contrast ratios. Token-based system enabled rapid iteration.

User Flow Comparison

Before: 7 Taps Across 3 Screens (18.3s avg)

After: 2 Taps from Home (4.2s avg)

Before vs. After Comparison

Before — Original Flow

7 taps across 3 screens to start meditating
Mandatory 10-question onboarding
No session save/persistence
Wheel picker for duration (no preview)
67% task completion rate
Avg. 18.3s to start first session

After — Redesigned Flow

2 taps from home screen to start
Optional onboarding (skip available)
Auto-save with resume prompt
Inline duration slider with live preview
94% task completion rate
Avg. 4.2s to start first session

5

Design System & Tokens

The design system uses a calming palette anchored in blue-green tones (associated with tranquility in color psychology) with Montserrat as the primary typeface for its clean, approachable character. All tokens are documented in Figma with usage guidelines.

Token	Value	Usage
Primary	#0EA5E9	CTAs, active states, links
Calm	#06B6D4	Session timer, breathing animations
Relax	#8B5CF6	Premium features, streaks, achievements
Background	#0F172A	Dark mode background (default)
Surface	#1E293B	Cards, sheets, elevated elements
Type	Montserrat	Headings, body, UI elements

Accessibility: All color combinations tested with Stark plugin and manual contrast verification. Dark mode palette achieves minimum 4.5:1 contrast for body text and 7:1 for critical UI elements. Screen reader testing conducted with VoiceOver (iOS) and TalkBack (Android).

6

Usability Validation & Results

The redesigned prototype was validated with 10 new participants using the same task scenarios as the original study. Results were measured against 5 key metrics aligned with HFI's effectiveness framework.

94%

Task Completion

4.2s

Avg. Time to Session

82

SUS Score (Excellent)

0

Critical Issues Found

Metric	Before	After	Improvement
Task Completion Rate	67%	94%	+27 points
Avg. Time to Start Session	18.3s	4.2s	77% faster
Number of Taps	7	2	71% fewer
SUS Score	52 (OK)	82 (Excellent)	+30 points
Critical Issues	5	0	100% resolved

Key Takeaways

Measure everything: Without baseline metrics (67% completion, 18.3s time), we couldn't prove the redesign worked. Quantitative data makes the case.
Fewer taps = higher completion: Reducing from 7 taps to 2 had a direct, measurable impact on task completion (67% to 94%).
HFI methodology scales: The heuristic evaluation framework works for any interface — not just meditation apps.
Design tokens enable iteration: A token-based system made it fast to test 3 different wireframe concepts without rebuilding from scratch.