Through this qualitative usability study on Apple Fitness+ (no association with Apple), I gained extensive practical experience running user interviews and usability tests, organizing large amounts of data, and translating interview and user data into usability recommendations. I also learned how to successfully navigate difficult team dynamics. This project was extremely formative for my career in UX.
For my Introduction to Qualitative Methods class during my Junior year, I worked with 3 teammates to evaluate the usability of Apple Fitness+. Our goal was to understand how first-time users of AF+ search for workouts, organize content, set fitness goals, and customize their experience, and to identify opportunities for Apple* to reduce friction within these core interactions.
*Note: we did not collaborate with Apple for this project. We used Apple Fitness+ as our topic to learn more about qualitative research and usability testing for our class project.
ROLE / Context
UX Researcher
Four-member course research team
DISCIPLINE
Qualitative UX Research
TIMELINE
Feb – Apr 2026
TOOLS / Topics
Figma, Google Docs, Google Sheets, Google Slides
User interviews / interview guide & question writing
User recruitment / survey writing
Usability testing / moderation / analysis
Affinity diagramming
Comparative analysis
Brainstorming, Interview Guide, Prep
Brainstorming / Problem Statement
After some course onboarding and introductory classes, my team started brainstorming ideas for what product we wanted to do research on. We were given a choice between products fitness like Strava or Peloton, planning a trip like Booking.com or Tripadvisor, or some sort of grocery shopping / food delivery product. We eventually landed on Strava, because we thought its social features would be a great research subject. However, another team ended up filling out their product preference survey before us and chose Strava, so we had to pivot to studying Apple Fitness+.
After the switch, we identified a problem to guide our research: Apple Fitness+ is on every iPhone, but wasn't even in the top five most popular fitness apps, even behind more specialized running apps like Strava. You would think people would at least try the default app, and that some percentage of the 1.5 billion of them would stick with it (DemandSage). We figured that if Apple understood how users perecive AF+ and how they utilize the type of guided workouts (asynchronous classes) offered on the app, they could better differentiate their app and be a more appealing choice.
Creating the Interview Guide - 02/19/26
For the first set of assignments, we created an interview guide, recruitment materials, consent forms, and screener survey. We also developed a recruitment strategy and script, and landed on formal research questions.
See a number of our interview guide materials below:
Problem Statement, Recruitment Materials, Interview Guide. Drag to see more
We gathered 48 responses to our screener survey, all between the ages of 18-24 and 25-34 and all iPhone users. Along with contacting other students, friends, and family members, we sent out our survey on the Michigan Law school mass email list—my sister was a second-year law student at Michigan at the time! This unique strategy got us a ton of potential participants!
The process of creating this document and doing in-class activities in small sections taught me so much about user interviews. For instance, I learned:
How to avoid writing leading questions
How to pace the questions in an interview by leveraging the funnel technique
How to write effective probes and dig deeper when a point of interest comes up
How to create these baseline materials and set up a user interview. I had never done this before, so this was all new and challenging, especially the interview guide itself!
Piloting, User Interviews, Affinity Diagramming & User Interview Report
Piloting and Interviewing - 02/15/26 and 03/08/26
After prepping and creating the materials, we moved into piloting and conducting our interviews. Each of us did one pilot interview, and three more interviews that we used in the dataset. I did my first ever user interview, and while it wasn't perfect, it went pretty well! I learned a lot, mostly about how effective interviewing takes more than a well-written question guide. You need to actively listen, be flexible with the guide (this one was huge for me), and be aware of how your reactions and responses can influence participants. Applying feedback from my instructors after the pilot helped me improve quickly and perform three quality interviews. Some more lessons I learned from this process:
You have to pilot
The pilot revealed some redundancies and spots where I found I wanted to dig deeper. I also cut down some questions so I wasn't reading them word-for-word. Piloting is a super important first step.
Follow the conversation
In the pilot, I found myself redirecting the participant towards other questions, because that was the order they were in in the guide. I learned how to gently guide a conversation and stay on a thought.
Get comfortable with silence
I learned just how long a few seconds can truly feel. I was very intentional about giving participants time to think, and I became more comfortable taking a moment to gather my thoughts.
This is a skill
This experience taught me to treat interviewing like any other skill. I gained a lot of confidence, learned to apply critique, and became more intentional about how I asked questions and responded to answers.
I'm pretty good at this!
The instructional team complimented my ability to summarize stories and develop probes grounded in what people shared. I continued refining this approach as I interviewed, searching for rich stories rather and not just relying on my question guide
Interview tracking spreadsheets (information redacted for privacy), interview data selections, final report selections. Drag to see more
Affinity Diagramming / Final Research Report
Team Issue
Now, it was time to sort the data and create a report with our findings. We used Affinity diagramming to achieve the first step—it's a great tool that allows you to visually organize and categorize, and was perfect to figure out themes and patterns in our data. We 9 participants' worth of data to sort. We had planned on having 12 for this activity, but one of my team members didn't hold up their end of the work. This was obviously a challenge — we had to continue on without that team member for this assignment (they weren't responsive), so I took on a decent brunt of their work. I also handled conversations with that team member and the instructional team to tell them what was going on. I used the appropriate channels to report our situation and personally reached out to the team member too. It sucked—but for right now, we just needed to get our data organized, even if we were missing a chunk of it.
Affinity Diagramming - 04/05/26
That being said, we still had so much data to look through! Our prep work of highlighting quotes and looking for them before we began diagramming (as we cleaned the transcripts) proved super helpful. We got to work. We used FigJam, and tried the AI organization grouping feature. We found it somewhat helpful, but really only used it to help jump start the real work. From this process, we discovered some themes and further organized ideas into categories within those themes. See the carousel above for a quick glance at what we eventually landed on.
I took a lot of initiative with this part of the work—going to office hours to really dial in the diagram, dealing with my team member (who was fine, and came back for the next assignment and contributed as expected), and spending my own time pouring over all of this so we would have a great basis for our final report.
Final Research Report - 04/05/26
Once we had our data organized, we put together a slide deck to communicate what we learned and what Apple Fitness+ could theoretically do with those insights. We created two personas to represent the types of users we saw through our interviews: someone newer to the gym who wants guidance, and someone more experienced who uses guided workouts to keep their routine flexible.
We then identified three key findings, using participant quotes to keep them grounded in what people shared in the interviews:
Participants valued learning proper exercise form from guided workouts
Social pressure could be both motivating and intimidating for participants to work out at all, and guided workouts provided a judgement-free space for them
The balance between music and instructor audio affected participants' workout experience
From those findings, we proposed three recommendations:
Real-time form feedback using AI that can display proper form and provide personalized corrections. Interviews showed that having good form is an important part of working out, and they wanted guidance to continue on their fitness journey.
Adding social accountability features like friend challenges, shared workout sessions, and leaderboards that are clearly visible in the app. Interviews showed that users stayed more consistent when they had teammates, a friend, or a group behind them and motivating them.
More control over the audio and mixing experience to give participants more control of the energy and vibe of their workouts
It was really cool to see the conversations we’d had with participants turn into actual design recommendations. Our instructors highlighted the clear organization of our affinity diagram and how both personas built off their supporting quotes, which was really encouraging after all the work we put into the analysis. This step taught me a lot about translating research into a clear story. Reporting what participants said, yes, but also explaining why it matters and how it could inform a better experience.
Comparative Analysis
Timeline / Wrinkles
You might have noticed that I included dates for all the milestones in the project so far. The winter semester runs from January through late April, with exams at the end of April and beginning of May. By the time we finished our research report, it was already April 5th, and our final work was due by the 26th. We had three weeks to complete a comparative analysis, full usability test, and a final testing report. Oh—and that's on top of finals and projects for three other classes (like my MMB website redesign and 311 data study), and marching with Redline Percussion. That was a particularly deep wrinkle, because I was traveling with Redline to Dayton, OH, for WGI World Championships from the 13th through 18th. Plus, I had a concert with the University Band the morning after on the 19th, and a camp with the Cavaliers Drum and Bugle corps the following weekend from the 24th through the 26th.
My life was pretty crazy during April 2026—making history with Redline with our competitive results at WGI Worlds, completing three large final projects simultaneously, preparing for drum corps… but I wouldn't have had it any other way! It was time to get creative and make it work, something my experience marching drum corps and indoor has taught me over and over. It took a lot planning, hard work, late nights… all the stuff… but nothing good comes easy!
Comparative Analysis
It was time to get to work. We identified three comparators: Peloton, a direct comparator; YouTube, an indirect/partial comparator; and Strava, an analagous comparator. Doing this analysis, we were able to identify a lot of issues with Apple Fitness+ and develop recommendations for how Apple might improve the app experience.
For example, you can't use any playback options besides playing or pausing, so we used YouTube and Peloton, two platforms that do this very well, to develop a suggestion to bring a full range of video playback features and to show real-time workout metrics overlayed on the video. Strava came in to play with the other two comparators when we talked about the lacking social features on AF+. In total, we developed five main recommendations:
Improving the playback experience when in a workout video
Adding full video playback control
Showing real-time fitness metrics (like burned calories)
Expanding and refining the audio customization / personalization
Robust choice of mix of trainer / workout audio
Support for more language / subtitle tracks
Audio hints clarified and turned off if not available
Enhancing social features
Let users add notes when finishing a workout
Add "like" and comment support
Allow users to react to / pin comments
Revamp the "Program" UI and placement
Advertise these more clearly
Detailed info presented upfront on a program summary page
Expand search filtering options
More filters
Allow combining multiple filters
Using these comparators helped frame our suggestions and give real examples of these features working in other apps that are stealing marketshare from AF+. See the full comparative analysis below in the Figma file:
Usability Test + Report
In-Class Usability Test - 04/04/26
This was basically some training before we did before doing tests for our individual projects. In this exercise, my team worked with multiple other teams to conduct a usability test of Gamma AI, an AI presentation and website builder. We wanted to understand how first-time users could generate a presentation, edit its content and design, export it, and turn it into a website. We created five tasks (with a backup task for insurance), with success criteria and estimated completion times. We also wrote pre- and post-test questions, a moderator script, and a participant-facing task list. Before handing everything off, we piloted the tasks with other students in the course, and adjusted the tasks based on feedback from the students and our instructional team.
I served as my team’s coordinator, helping communicate with both the other task design team we worked with directly, as well as the live testing and data logging teams who performed the usability test in live time. While the÷ other teams handled moderation and analysis, our role gave me experience with the preparation that makes a usability test possible. This was a different side of research from interviewing. Instead of simply asking people about their experiences, we designed opportunities to see how participants actually used a product, and the struggles that go along with that. I learned about defining clear task outcomes, planning for situations where a participant might get stuck, and coordinating research across multiple teams.
Usability Test Planning - 04/23/26
After learning about participants’ guided workout routines through interviews, we shifted to using usability tests to gather more concrete data about how users actually experience AF+ specifically. We created a usability test plan focused on finding workouts, building and saving workout routines, and customizing video and audio playback. I wrote many of the tasks and their expected paths and success criteria, helped define our participant criteria, and worked on the consent form and post-test questions. I also visited office hours and worked with the teaching team to refine the tasks and our facilitator guide. A big part of this was learning how to give participants a clear goal without explicitly telling them which buttons to press, and letting the interface play itself out.
We piloted the plan before conducting the official tests, and subsequently revised the task wording, order, and expected paths based on that experience. For example, the pilot revealed that browsing and filtering a broader category was a valid way to find a specific cycling class, rather than using search like we expected users to. This reinforced what I learned from interviewing: it's great to have a plan, but it's better to be flexible and see what participants do when given no instruction. It can tell you a lot about their attitudes and the design of a system. I learned tons about preparing usability tests that provide enough structure to collect useful data while still leaving room to see how people naturally navigate from doing this portion of the project.
Usability Testing & Final Report - 04/26/26
For our final assignment, we conducted 12 in-person usability tests, with each of us moderating and data logging three sessions each. I performed the tests with my three participants in their homes, and later reviewed the recordings to log task completion, hesitation, errors, comments, etc. Something that really stood out was that completing a task didn’t necessarily mean the experience was good. Participants could eventually find a feature, but still feel really frustrated by confusing labels or unexpected navigation. I also noted mistakes in my own testing; small mistakes in the testing like partially resetting the system might have influenced the data in some way. Overall, I felt very good about the work I did with these tests, especially going from 0 to 100 on usability testing in only a couple of weeks.
My team and I compared patterns across our logs and prioritized four findings for the final report: confusion around the “Workout” tab, confusing “Stack” terminology (AF+'s choice for the more common "queue" to play multiple videos in sequence), hard-to-find search and filtering controls, and limited visibility of goals and planning. From these, we developed recommendations for clearer labels, more visible navigation, and refocusing the Summary page to center on goals and personalized progress. I found it very rewarding to connect specific moments from my sessions to patterns across the team’s research. This brought the project full circle, as our interviews helped us understand what people valued in guided workouts, and usability testing helped us identify what could get in the way of accessing that value in Apple Fitness+.
Final Thoughts
I learned how to…
Recruit for, pilot, and moderate user interviews
Clean a transcript and organize interview data (with affinity diagrams)
Conduct a competitive analysis
Moderate a usability test, interpret usability test data
Turn raw data into findings and recommendations
Unique Challenges
Team issues
Team member not conducting interviews and didn't do their work. Handled professionally with teaching team and helped them get back on track.
Juggling this study with other classes and marching indoor percussion and drum corps
Many due dates in April while WGI finals and drum corps rehearsal camps were happening. I got creative and made it work!
I had never done qualitative research before!
Everything listed above was brand new to me — but we came out the other side with a great project!