Spot Me
At a glance
Many lifters start casually without guidance, often risking injury as they increase weight with poor form. Spot Me helps with solo workouts by using AI-driven pose estimation to "spot" your form. It gives real-time feedback, so you can focus on lifting your best without the worry of getting hurt.
Principle Designer
4 months
Mobile App
Desk research
A growing trend in health tracking and digital fitness.
This has been fueled by a pandemic-driven shift towards home workouts and health-tracking devices. While the surge in digital fitness has slowed as life returns to normal, the trend persists, with many people still dedicated to tracking their health and investing in fitness technologies.
There were 850 million downloads for fitness apps in 2023, indicating a turn to digital fitness.
Fitness app revenue increased in 2023 even as usage and downloads declined, as more users stayed with their subscriptions.
User Interviews
How do people at different fitness levels work out?
I conducted 7 discovery interviews to gain a holistic understanding of how people at different fitness levels incorporate exercise into their lives. I focused on why they started working out, their current routines, and what motivates them. This approach allowed me to identify 3 meta insights, which informed my ideation process.
01
02
03
NICOLE
CARDIO
Following a half-marathon plan.
If she derailed from the plan, she’s not sure how she would be able to get back to where she was.
ANGELIKA
Weights, cardio
Followed an intense HIIT program since she began working out but recently her goals changed. Now, her focus is on building her strength and muscle.
DENNIS
Weights, Cardio
"After a while, you kind of get addicted to the data. Whenever you go through the same data, I think did I go faster this time?"
ALEC
CARDIO
"The first thing I do after I run is check Strava. It’s rewarding to see the stats or see something you did."
VERA
Weights, Cardio
"I think using the right muscle is really important for yoga or weight training because a wrong pose could cause a lot of injuries."
HANA
Weights, Cardio
"At the time, I wasn’t sure if my pose is right because no one is there to check my accurate pose."
Divergent Solutions
Turning main insights into key features.
I developed four divergent ideas to test with users, and their feedback revealed a strong preference for solutions focused on guidance. All participants expressed a desire for the benefits of a personal trainer- but found these services unaffordable.
1
Quiz generates daily workout routine
2
Biometrics alerts for abnormal metrics
3
360° viewable poses for understanding form
4
Smart mirror, visual feedback
Personalized Workout Program
A long-term program provides weekly routines to help users reach their main goal. Leverage 'streaks' to motivate ongoing progress.
Visualize Which Muscles to Target
Tutorial training videos should visually indicate what muscles to target and improve users' ability to accurately teach themselves the correct form.
Monitor User Biometrics for Safety
Highlight user biometrics to track progress and monitor if they’re pushing themselves too hard or not enough, helping prevent injury.
Prototyping & Testing
Guiding users through their workouts to ensure safety.
This concept focused on walking users through each set by visualizing which muscles to target. It leverages wearables to track biometrics and warn users if they are pushing themselves too far. I tested the mid-fidelity prototype with 6 users dedicated to using fitness apps and working out consistently.
And then I realized the real problem —
Solo weightlifters struggle to teach themselves the correct form, increasing the risk of injury.
All 6 weightlifters I spoke with had experienced severe injuries from pushing too hard while lifting alone, unaware their form was incorrect. Many skip formal instruction and rely on watching others or going solo, which leads to poor form and wrong positioning.
Going back to Research
Is the solution technologically feasible?
I revisited the research phase to explore existing technologies that could support my proposed solution. Based on insights from interviews with weightlifters, it was essential to assess how well the solution aligns with their current lifting practices.
Findings & Iterations
Tailoring the user experience to weightlifter's needs.
With my area of opportunity and target users defined, I moved on to testing my high-fidelity prototype with 5 weightlifters. This process led to several key iterations, refining the UI and UX to better align with their workout habits and seamlessly integrate into their routines.
Guided Workout Sets
Before:
The progress bar and timer indicate the user's position in the workout and how long the set should take. A bottom sheet swipes up to view the full workout routine.
After:
The number of reps and names is emphasized instead of timers, as users prioritize quality over speed. 'Spot Me' toggle starts video analysis for form correction.
Form Correction Alert
Before:
When AI detects incorrect form, a temporary modal card provides in-depth feedback. However, users indicated they'd prefer a shorter copy during the workouts so that they could react quickly.
After:
Bite-size feedback is delivered visually in large, clear letters and audibly through a coach's voice, ensuring accessibility for users who are color-blind or distant from the screen.
Post-Workout Analysis
Before:
Testers wanted to see clear statistics immediately after the workout. Better if it could compare to past stats.
After:
Statistics on the user’s accuracy improvements are clearly spotlighted and celebrated.
The Final Solution
Spot Me
An AI-driven fitness app that corrects your form in-real time so that you can keep the focus on lifting at your personal best.
Core Features of the Final Experience
Find your daily workout and track this week's progress.
Progress is tracked with check marks to indicate their streak, while completed workouts display accuracy.
Coach-led video tutorials guide users through sets.
Sets are broken down into the required weight and number of reps and include a video demonstration on how to perform the exercise correctly.
'Spot Me' mode activates video analysis, alerting user of incorrect form.
The modal disappears once the adjustment is detected and the position is determined safe.
Post-workout analysis of form accuracy.
An easily digestible breakdown of the user’s form accuracy. Highlights statistics on their progress and provides further guidance on corrections made during that day’s workout.
Impact
of users would trust AI recommendations.
users had been injured due to incorrect form.
users had been injured due to incorrect form.
What Did I learn?
Design Isn't a Linear Process
At the start of this project, I was uncertain about which fitness problem to tackle. It wasn’t until my mid-fidelity interviews that I focused on weightlifting. The process involved revisiting my research, but that’s the thrill of design— drawing, researching, testing, and repeating. Keeping the user’s needs as my north star guided me to a genuine area of opportunity.
Don't Forget About The Tech!
I love getting into UI and UX design, but that’s just one piece of the puzzle. Researching AI has shown me how crucial it is to stay updated with the latest tech to keep innovating. I think it’s all about balancing the tech we have with imagining what it could be to keep pushing the boundaries.