UX Design Internship · Google Fitbit
Google Fitbit
Celebrating premium users' workout achievements within the Fitbit Ecosystem.

Role
UX Design Intern
Timeline
3 Months
Team
Fitbit Service UX Team
Tools
Figma, internal metrics tools
THE BRIEF
Project Context
As a UX Design Intern on Fitbit's Service UX team, my mentor handed me a focused brief: design a feature that helps Fitbit Premium subscribers understand and celebrate their workout achievements.
Premium users already trust Fitbit with their health data. The challenge wasn't collection — it was meaning-making. How do you transform raw workout logs into something that actually motivates people to keep going?
My Role
- •End-to-end ownership of the solution space: from concept exploration to engineering handoff
- •Research with Fitbit expert coaches and usability testing with 8 real Premium users
- •Cross-functional collaboration with PM, Engineering Lead, and UXR team
DISCOVERY
Understanding the User
Sue
Fitbit Premium User
"I want to improve my health, not study a spreadsheet. I need motivation, not just math."
Goals
- •Wants to stay motivated but gets overwhelmed by raw data.
- •Needs clear progress tracking without cognitive load.
AS a First-time user
Needs to understand how the feature works before engaging.
AS a return user
Needs to see weekly progress details
The design needed to cater to both discovery (learning) and retention (monitoring) phases.
THE CHALLENGE
Logic vs. Emotion
Fitbit collects millions of data points, but raw numbers feel cold. The design challenge was to transform 'quantification' into 'celebration' without breaking the medical trust of the brand.
Users engage with fitness data logically, but they stay for the emotional reward. We lacked the 'spark' of celebration.
DATA
REWARD
IDEATION
Exploring Engagement Models
I didn't start with one solution. To avoid anchoring too early, I explored 4 distinct engagement models — each testing a different hypothesis about what would actually motivate Premium users.

Daily Plan
Too prescriptive. Users reported feeling anxious about 'failing' daily tasks.

Emotional Feedback
Too subjective. Lacked the actionable, hard data insights users came for.

Gamification
Fun but too casual. Diluted Fitbit's professional authority as a health coach.

Insight Card
The sweet spot. Combines data accuracy with professional guidance.
HOW WE DECIDED
Converging on a Direction
After the exploration phase, I ran a working session with my mentor to map out the trade-offs of each model. Daily Plan felt too prescriptive — users reported anxiety about 'failing.' Emotional Feedback was too subjective. Gamification diluted Fitbit's authority as a health platform.
The Insight Card was the only model that held both sides of the equation: the data rigor Fitbit users trust, and the sense of progress they actually need to stay motivated.
RESEARCH & CONSTRAINTS
From Direction to Data: Two Hard Problems
Selecting Insight Card gave us a direction — but not a design. Two challenges stood between the concept and a real solution: I needed to identify the right metrics to visualize, and I needed to understand what data was actually buildable.
The Science Behind the Metrics
To define what a 'balanced' workout actually means, I conducted interviews with Fitbit's internal expert coaches. I learned that the U.S. Department of Health and Human Services (HHS) defines holistic fitness through three physiological pillars — giving the feature a scientifically credible backbone.
"Most users focus on cardio, but true fitness requires balancing strength training with high-intensity intervals to trigger metabolic change."
— Head Coach, Fitbit

The workout triad grounded in HHS fitness guidelines.
Working Within Engineering Constraints
Different workout types capture different data fields — a running session and a yoga class don't log the same metrics. Engineering needed a balance calculation that worked consistently across all activity types, using only the data Fitbit reliably tracks. I developed two proposals with explicit calculation logic for the team to evaluate.
Proposal 1
Triangle Area Ratio
Calculate balance as a geometric ratio: compare the area of the user's actual workout triangle (S1) to the theoretical maximum equilateral triangle (S2). The ratio K = S1/S2 produces a single balance score from 0–100%, classified as Low (0–30%), Medium (30–70%), or High (70–100%). Simple, continuous, and directly tied to the radar chart shape.
Proposal 2
Per-Metric Bucketing
Independently classify each of the three metrics (Cardio, Strength, Intensity) into three levels: Low, Medium, or High. Combine them as a composite state (e.g., [High, Low, Medium]) to surface specific gaps without producing a black-box score. Transparent, granular, and no special scoring required.


Both proposals were documented with full calculation logic and delivered to the engineering team for feasibility evaluation. Given the 3-month internship timeline, the final implementation decision extended beyond my tenure — but both were referenced at a cross-functional Product Area level as Fitbit considered how to surface workout guidance and achievement across the product line.
THE SOLUTION
The Insight Card
The three HHS-grounded metrics — Cardio, Strength, and Intensity — became the backbone of the design. The challenge now was to make balance legible at a glance, without overwhelming users with data they didn't ask for.
The Insight Card translates workout data into a radar chart that lets users see their balance shape instantly. A progressive disclosure pattern — glanceable summary to detailed breakdown to trended view — ensures the depth is always available, but never forced.

Progressive Disclosure
Designed a glanceable 'Insight Card' as the entry point, providing a high-level summary before revealing deep-dive metrics. This prevents data overwhelm.
Intuitive Visualization
Created a custom radar chart component that transforms abstract 'balance' data into a clear geometric shape, allowing users to spot gaps in their routine instantly.
Native System Alignment
Applied Google Material Design tokens (typography, spacing, and chip components) to ensure the feature felt like a native, seamless part of the Android ecosystem.
VALIDATION
Validating with 8 Fitbit Users
Working with the UXR team, I recruited 8 real Fitbit Premium users to walk through the full prototype flow — from first-time onboarding to the glanceable card, detailed breakdown, and trended view. The concept scored an average of 8/10 in desirability. But the more important outcome was what the testing revealed about what still needed to change.
User Quotes
"The well rounded approach to fitness is definitely a goal for me..."
FL, Teacher, 35-44
"Duration is a helpful qualifying metric... but what else speaks to quality?"
MM, Doctor, 60+
"Long term views are a strong desire for many of us long term Fitbit users. I have a database of my 6 year trends pulled outside of Fitbit."
TT, Runner, 35-44
Key Findings
Motivation
4 out of 7 users found the balance concept organically motivating.
Clarity
Scores improved significantly from first-time view to return-user view.
Desire for Trends
Users expressed a strong desire for long-term (monthly/quarterly) views to see progress.
Iterating Based on Feedback
The clearest signal from testing: users wanted to see progress over time, not just a weekly snapshot. I added two new views in response.

Progress Summary
A monthly calendar view showing workout consistency over time — directly addressing users' desire for long-term trend visibility.
Week View
A weekly breakdown giving returning users the detailed, day-by-day data they requested during testing.
Both additions were validated against the original desirability scores — clarity improved significantly from first-time to return-user experience.
OUTCOME
From Intern Project to Design System Asset
8/10
Desirability
Average score across 8 real Fitbit Premium users in usability testing.
4 of 7
Organically Motivated
Users described the balance concept as motivating without any prompting.
Pixel Watch
Cross-Product Adoption
The Insight Card design pattern was adopted beyond mobile to Fitbit's wearable ecosystem.
"Joyce showed strong leadership skills and technical and analytical abilities — synthesizing various data collection requirements and constraints, and delivering two recommended solutions with detailed calculation logic to be referenced on a cross-functional PA level."
— Boriana Viljoen, Senior UX Designer (Mentor) · Google Fitbit

My amazing mentors and me at the Fitbit SF office.
What I Took Away
Fail fast, iterate often.
I spent too long in the diverge phase with no clear exit criteria. What I learned: the goal of exploration isn't to find the perfect idea — it's to collect signal fast enough to make a confident decision.
Engineering constraints sharpen design.
Fitbit's data limitations forced me to define 'balance' mathematically, and that rigor made the solution more credible with engineering. My software background made it possible — and it's now central to how I work.
Next Project
Leah — AI-Powered Experience Design