Year:
Purdue UX Design Studio Spring 2025
Overview
Project overview
For this project as a part of Purdue's UX Design Studio, our team set out to rethink how fitness trackers support motivation and habit building. We focused on college students using the Apple Watch and found that static goals often led to guilt, stress, or even “cheating the system.” To address this, we designed an adaptive goals feature that adjusts daily activity targets based on how the user feels mentally, physically, and with their schedule. Our solution helps students stay active without the pressure of rigid goals, building healthier long term habits while keeping trust and motivation at the center of the experience.
Click to view - Fig jam
Click to view - Documentation Master Copy
Project Stages
Investigate issues related to body data tracking, mental models, and behavior change.
Interview and observe users.
Analyze data using appropriate methods, identifying and documenting key findings that inform our problem framing and design direction.
Mockup solutions iteratively, starting with low and medium-fidelity materials and creating multiple versions along the way.
TEAM 4 MEMBERS
Ani Berry – UX Design & Psychology
Emily Clark – UX Design
Sarah Neumann – UX Design
Priscilla Tam – UX Design & Computer Systems Analysis
Starting with Research
To understand our challenge more deeply, we kicked things off with secondary research to explore people’s behaviors, motivations, and frustrations with fitness trackers. Our main goal? To learn what actually gets people to use these devices, and more importantly, what keeps them coming back long term.
Questions That Guided Us
We wanted answers to a few key questions:
What motivates people to stay active, internal factors like health and confidence, or external ones like streaks and social features?
Which body metrics are most important or confusing to users?
What kinds of goal setting systems actually help people stick to routines?
Why do people stop using fitness trackers, or cheat the system?
We gathered insights from academic sources and articles and summarized them in a Fig jam board for the team.
What We Found
Motivation
Studies showed that intrinsic motivators, like wanting to be healthier, feel better, or gain confidence, led to more consistent fitness habits than extrinsic ones like rewards or leaderboards. Personalized support and social enjoyment also played key roles.
Body Data Tracking
Trust and clarity were big issues. Users often doubted the accuracy of data like steps, calories, and sleep. When the data didn’t match their intuition, they got frustrated. People also valued different data at different times, so trackers need to adapt to each person’s changing goals.
Mental Models & Goal Setting
Smaller, realistic goals were more effective over time, especially when they adapted based on user progress. Studies showed that personalized, daily step goals actually increased physical activity and enjoyment.
Behavior Change
Features like reminders, self tracking, and feedback were all important for forming habits. But competition could backfire, some users felt discouraged or even cheated to keep up. So balance and flexibility are key.
The Downside of Tracking
Tracking can also lead to stress. Some users focus so much on numbers that they ignore how their bodies actually feel. Social sharing features sometimes caused unhealthy comparisons. And many apps overlooked partial effort, if you didn’t fully complete a goal, it felt like you failed entirely.
Key Takeaways
From all this, we identified a few core needs:
Help users stay intrinsically motivated
Adapt goals to fit changing needs and energy levels
Build trust in the data
Reduce guilt and stress by supporting flexible use
Exploring the Context
Mapping It Out with Activity Theory
To dig deeper into how fitness trackers fit into people’s lives, we used an activity theory framework to break down the experience of using an Apple Watch. We looked at the relationships between the user, their tools (the watch, app, and data), and the outcomes they were trying to achieve.
Pain points / What We Noticed
Not everyone shares the same fitness goals
The watch isn’t convenient for all users
Some users don’t fully trust the data
It can be hard to find a real sense of community
Framing the Problem
With those pain points in mind, we narrowed our problem statement:
People often use fitness trackers to earn rewards or close rings, but they’re not always actually getting healthier. Many lose motivation or find ways to game the system instead.
We then reframed the challenge as a design question:
How might we make fitness goals more meaningful and motivating, so users want to stick with them, not just beat the system?
Early Ideas
Once we had a clear direction, we brainstormed ways to reimagine the fitness tracking experience.
Screenshot of idea board
*Link to FigJam
Concepts We Explored
Locking social media apps until users meet their fitness goals
A “rival mode” to compete with your past performance
A college campus based network for group fitness motivation
A feature that adapts goals based on your mood, stress, or energy levels
We then brought in AI (Claude and ChatGPT) to help test our ideas and explore feasibility.
AI Recommendations
Claude suggested reframing fitness as a personal story, with milestones, reflection, and intrinsic rewards.
ChatGPT recommended combining social fitness circles with “adaptive streaks” that change based on how users feel.
After discussing with our team and getting feedback from our desk critique, we decided to focus on adaptive goals for college students.
Deep Dive: Adaptive Goals
Why Adaptive Goals?
Most fitness apps use static goals based on general info like height and weight, but not on how someone actually feels or behaves.
Research showed that:
Adaptive goals help users maintain progress more effectively
They lead to better outcomes over time (Adams et al., 2017)
Apps like MyFitnessPal still don’t offer truly personalized goal adjustments
This backed up our idea to build a smarter system, one that flexes with a person’s daily experience.
Knowing Our Users
Interviewing College Students
We talked to peers who use Apple Watches to better understand their experience. A few themes stood out:
Most used their watch to track steps, workouts, heart rate, and to close their activity rings
Closing rings made them feel accomplished, but missing a day led to guilt
Busy days (like back to back lectures) made it hard to stay active
“Adaptive goals” were well received, people liked the idea of goals that adjust based on how they feel
Some worried it might make them too lenient and reduce motivation
(Interview questions are available in the appendix of the final documentation)
Validating with Secondary Research
We also looked at broader data about college students’ fitness behaviors:
Activity levels tend to fluctuate with the school calendar (Bai et al., 2020)
Students are more active on weekdays, less on weekends
Exercise has strong links to mood, mindfulness, and healthy habits
Physical activity is a major stress reliever for students
This confirmed that adaptive goals would be especially helpful for this user group, because their schedules and energy levels are constantly changing.
Designing and Iterating
Evaluating the Current State: Where Could Our Solution Fit?
To figure out how our proposed Adaptive Goals feature could integrate naturally into the Apple Watch experience, we started by conducting a current state evaluation. This helped us understand how the system currently works, what gaps exist, and where our feature could genuinely enhance the user experience without disrupting the flow Apple users expect.
What We Learned About the Current System
Apple Watch users can adjust activity goals manually, but it’s a static and user initiated process
There’s no built in support that accounts for fluctuations in energy, emotions, or daily schedules
The “Pause Your Rings” feature exists, but it functions separately from goal setting and doesn’t support more nuanced needs
There’s no middle ground for users who might be stressed, tired, or overwhelmed, yet still want to stay somewhat active
The system doesn’t ask about the user’s current state or feelings, and there’s no adaptive behavior based on personal context
Although the phone app includes trend graphs and scheduling, it lacks any emotional or mental check in features
How Our Solution Complements the Existing Flow
To address those gaps, we created a low-fidelity prototype of a simple daily check in. It asks three quick questions focused on physical readiness, mental focus, and schedule constraints. The responses feed into the Adaptive Goals feature, which dynamically adjusts the activity rings to reflect how the user is feeling that day.
This check in flow was designed to feel as natural as possible within Apple’s interface minimal, purposeful, and user centric. We also made sure that the check-in process didn’t feel like an extra burden. Instead, it’s a moment of reflection that informs smarter, more empathetic tracking.
We later refined this idea through mid-fidelity mockups, focusing on integrating seamlessly into the existing Apple Watch ecosystem. We paid close attention to maintaining Apple’s clean aesthetic, carefully choosing colors, language, and flow to avoid adding friction or confusion. Our goal was to make it feel like this feature had always been part of the Apple Fitness experience.
Our Solution: Mid-Fidelity Prototype
Our final concept introduces an “Adaptive Goal” option as part of the existing goal-change flow. After selecting this option, the user is presented with a quick overview of how Adaptive Goals work, followed by a brief questionnaire:
How are you feeling mentally today? (Focused, Neutral, Overwhelmed)
How are you feeling physically today? (Active, Neutral, Resting)
How busy are you today? (Busy, Moderate, Light)
In our mid-fi designs, we replaced emojis from earlier mockups with simple circular buttons to align more closely with Apple’s brand guidelines.
Based on these inputs, the Fitness app generates an adjusted Move, Exercise, and Stand goal for the day. Over time, if a user continues using Adaptive Goals, the app starts to learn from their behavior and adapts automatically. For example, if someone reports feeling overwhelmed multiple times and struggles to meet their rings on those days, the app may begin reducing goals in advance when that same pattern emerges.
To help users stay motivated, even on low energy days, we added supportive progress reports. These are designed to reinforce that every bit of movement still counts. The feedback is positive and motivational, helping users feel recognized for their effort, not penalized for falling short.
Behind the Scenes: How Adaptive Goals Work
From early research, we learned that many users have trouble trusting fitness trackers, especially when they don’t understand where the numbers come from or why they change. Studies also show that transparency in algorithmic decisions builds user trust and leads to more consistent usage.
That’s why we were intentional about explaining how our adaptive system works.
The feature relies on a machine learning algorithm that pulls from three primary data sources:
Self reported wellness data from the daily check in
The existing Apple Watch daily Move goal
A history of the user’s activity patterns from previous days
The algorithm evaluates these inputs to set a realistic and meaningful goal. For instance, if someone reports low energy and their historical data shows a trend of missing goals under similar conditions, the algorithm will lower the goals to keep things achievable. It adapts to behavior over time, aiming to meet the user where they are, not where a standard model thinks they should be.
Looking Ahead: What’s Next?
With just three weeks to complete this project, we weren’t able to fully validate or test every part of the system. There are clear areas we would explore further in the future:
Test with real users to see how Adaptive Goals affect motivation and consistency
Consider offering presets so users don’t have to complete a check-in every single day
Explore ways to increase long term engagement while minimizing friction
Evaluate feasibility and value with real stakeholders
Continue refining visual design to match Apple’s style, limited now due to Figma constraints
Final Feedback: Where We Can Improve
After presenting our concept, we received thoughtful feedback that helped us identify a few important questions and possible gaps in our solution.
Machine learning limitations
Some people raised concerns about how well machine learning can adapt to daily feelings. Since it learns from patterns rather than one-off emotional inputs, they wondered how effective it would be at truly personalizing daily goals.
Notifications and motivation
Others pointed out that notifications might actually hurt motivation if not handled carefully. Could the system reduce or time these prompts more effectively based on user behavior?
Ethical concerns about adaptation logic
We were asked whether the system could unintentionally reward low moods with less movement, potentially reinforcing negative patterns. It’s a valid concern and one we’d explore further with behavioral scientists if taken into development.
Overlap with existing features
Some questioned how this differs from simply using the manual goal setting tools already in the app. While our system adds adaptive intelligence, we’ll need to clarify that value more in future iterations.
These questions helped us sharpen our thinking and highlighted the importance of ethical design, transparency, and clear user benefit.
Team Contributions
Ani Berry
Led research and literature reviews
Conducted UX interviews
Designed initial prototypes and final walkthrough video
Helped shape the feature flow and documentation
Used ChatGPT for ideation and phrasing refinement
Emily Clark
Led desk and secondary research
Helped define the problem and structure the team’s process
Managed group coordination and documentation
Collaborated closely on final language and rationale
Sarah Neumann
Conducted UX interviews
Owned the current state evaluation and analysis
Helped refine storytelling for presentation slides
Used ChatGPT for summarizing feedback and streamlining insights
Priscilla Tam
Created diagrams like the activity theory triangle
Framed the core problem and led user journey mapping
Used Claude and ChatGPT to evaluate ideas
Wrote key content for slides and documentation
Use of AI in Our Process
Throughout this project, we used large language models like ChatGPT and Claude to support brainstorming, research synthesis, and language refinement. They helped us identify themes, reframe our concepts, and tighten up our writing for clarity and tone.
However, we were careful to verify all academic citations and research findings manually. AI tools were helpful accelerators, not sources of truth. Their biggest value came in helping us distill insights and communicate ideas more clearly within a tight timeline.
