Redesign Challenge Creation
Summary
About
“If Instagram had a feature for goal-driven communities, that’s Napoz.”
Napoz helps users create and join challenges that make them feel supported, seen, and motivated while doing it - building habits, sparking community.
Challenge: noun.
Once it was a call to competition - a test of strength. For Millennials, it became a tool for self-improvement. For GenZ, it’s something entirely new: creativity, community, identity.
Yet most challenge-based products today remain one-dimensional — built for fitness. What if a challenge was sketching every day, cooking once a week, or flexing your Wordle streak — and doing it with friends?
NAPOZ is redefining the challenge for the next generation — turning everyday habits into shared rituals.
This is more than habit-tracking. It’s a platform for personal expression, social motivation, and playful accountability, made social.
Any challenge. Any goal. Shared.
Create Challenge Flow
The Problem
Creating a challenge should feel exciting—
but for many users, it felt like filling out a form.
Despite being the heart of the product, 80% users don't know how to set up a leaderboard in Create Challenge Flow. Key pain points identified from data & user interviews suggests several usability barriers:
Decision fatigue from too many micro-choices
Low info clarity due to repetition and unclear wording
High interaction costs with multiple open-text fields
The Engagement Funnel Break down the full user journey into key moments: Onboarding → Discovery → Creation → Interaction → Return loop. "Creation" activity is less active than "Discovery" activity.
Interest exists in CCS(Challenge Creation Screen), but it was where most experienced confusion and drop-off. From there, I further narrow down the issues in leaderboard Setup.
Ideation
l led a cross-functional workshop with the founder, engineers, and marketing team. We aligned on a new IA that prioritized three sections in the creation flow: Challenge Basics, Tagging(a new PRD item to capture trend data for growth insights), and Leaderboard Setup(LS).
To improve leaderboard adoption—a key driver for monetization in the next product phase, I auditing 60+ user-generated challenges to identify the minimum viable inputs.
which was simplified into three core questions:
What's the data source?
How to calculate it?
What's the sorting order?
I then translate them into user inputs:
Added multiple choices to replace "Point Rules" text input.
Fields that are about fine-tuning were grouped under "Advanced Settings" that came last.
But it still felt like… a long list. So the next question is "How could I preserve customization while removing complexity for LS?" I explored three design options:
#1 One-Page Scroll
✓ Visually restructured to make inputs easier to digest
✗ Could still feel long
#2 Dynamic Preview
✓ Builds confidence
✗ Requires new components for Dev team
#3 Guided Flow by Challenge Types
✓ Supports progressive disclosure
✗ Force a choice too early & multiple screens
✗ Requires backend complexity to scale up
User Testing
We moved forward with option 1 as our direction. To validate this approach, I conducted both moderated and unmoderated prototype testing with returning beta users. The objective was to evaluate whether the restructured layout made CC(Challenge Creation) more simple and intuitive.
Users appreciated the streamlined structure, but many still experienced decision fatigue, especially during the Leaderboard Setup. The layout wasn’t the barrier—cognitive effort was.
This led to our next question:
What if users didn’t have to fill everything in themselves?
→ That’s where the AI-powered assist comes in.
AI Demo
I proposed a bold dream: users would simply type a few keywords, and AI would generate the challenge title, description, leaderboard settings, and even the cover image.
I collaborated with engineers, and within one week, built a working demo and presented to the founder. While image generation wasn’t feasible yet, the demo was proved to reduce user effort significantly through second round of user testing. We aligned on moving forward with text-based AI assist in the next release.
UI Exploration
While refining the product direction, I partnered with the marketing team to revisit NAPOZ’s brand palette. The original blue felt too safe for a product built around expression and motivation.
Purple finally won as the new primary brand color. It strikes a balance between individuality and boldness, evoking both creativity and emotional warmth—perfectly aligning with our product value. I also added a touch of gradient in the final UI like the last one since it was the #2 favorate.
The Solution
The final solutions concentrated on on simplifying the challenge creation experience while preserving flexibility for power users.
Key improvements included:
A restructured flow divided into Challenge Basics, Tagging, and Leaderboard Setup
Clear multiple-choice inputs replacing open-text fields
An AI-powered assist that generates titles, descriptions, and leaderboard settings from user-provided keywords
Advanced settings tucked away
Results
Task success rate 267% ↑
Task completion time reduced to 50%
Active users per screen 40% ↑
Setup AI features framework and further applying it into post creation flow
