Project Title

🧩 Challenge

With wearable tech becoming increasingly performance-focused, a company approached us to validate an unreleased wearable’s ability to estimate metrics used by runners and coaches to evaluate performance, injury risk, and running economy.

 

The Question:

Can this new device deliver lab-level insights in a real-world training environment — and do users actually trust the data?

Before launching the product, the company needed insight into both the data accuracy and the user experience of setup, wearability, and interpretation.

​

Project Timeline: 

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​​​​​​​​​​​​​​🧪 Solution

We conducted a comparative research study between the prototype and a benchmark wearable to assess agreement

Beyond data, we ran alpha testing, A/B validation, and intercept usability interviews to uncover technical inconsistencies and real-world friction points — all of which shaped the design of Version 2.

​

​

​

 

 

 

 

 

 

 

 

 

​

​

 

🎯 Research Goals

To help the product team prepare the wearable for public release, we set out to:

• Compare the prototype’s data to a known valid device across multiple running speeds
• Identify any accuracy limitations based on sensor placement and motion capture
• Understand runners’ experience setting up and using the device in real time
• Provide usability recommendations for refining the product pre-launch

​

🔬 Research Methodologies

As lead researcher, I designed a hybrid protocol integrating lab-grade biomechanical testing with in-the-moment user feedback

 

​

​

 

 

 

​​👟 Wearable Device Testing

Individuals completed laboratory testing using two devices 

​

🧑‍🔬 A/B Analysis + Alpha Testing

We ran both devices side-by-side during testing and evaluated validity and pain points such as:
• Agreement of outputs
• Sensor drift or failure events
• Issues with syncing, mounting, or movement artifacts

​

💬 Intercept Usability Interviews

After testing, participants gave immediate feedback:
 

​

 

 

 

​

​

🔍 Insights

✅ Comparable Output, Inconsistent Agreement
While the prototype showed no significant difference in average metrics compared to the benchmark; however speed had an effect on sensor accuracy and movement

​

⚠️ Sensor Placement Affected Data
Device placement introduced variability. Users often misaligned sensors despite instructions — a key UX opportunity.

 

😐 Low Trust in Metrics
Participants struggled to interpret device & app metrics. They wanted more context:

“Is that good? What does that even mean?”
“I wouldn’t know if I was improving without a coach explaining it.”

 

👎 Setup was Cumbersome
Users took longer than expected to attach all three pods correctly. One participant said:

“If this was part of my everyday warm-up, I’d give up.”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

​

​

​​​​

🧠 Concept & Application

Our findings informed both hardware and software redesigns. For the next version, we recommended:

• Simplifying device placement and auto-detection of positioning
• Adding in-app onboarding flows with visual cues
• Reframing stiffness data into interpretable language 
• Providing feedback in real time, not post-hoc, to support pacing and form correction

​🔄 Impact on UX

This project blended wearable performance testing with user-centered research to uncover both algorithmic blind spots and experience gaps.

Accurate data is only meaningful when:

• Users understand what it means

• The setup is seamless and trustworthy

• It aligns with real user goals (better form, reduced injury, personal progress)

By grounding the redesign in both physiological validation and user feedback, we helped shape a more intuitive, credible product that runners could rely on.