FlexMode logo with stylized blue stars.

Stay on track, no matter the setback.

Project Overview
Over 4 months, I was the lead ux/ui designer for a new adaptive workout feature for MyFitnessPal, addressing a critical user gap:

no existing modification support for users managing injuries or mobility limitations

Smartphone displaying a fitness app with options to select areas for modifications such as hips and knees, with checkmarks indicating selections, placed on a wooden surface with a blurred potted plant in the background.

(Above) Final prototype of FlexMode

A 2019 medium case study found

53% of users
want rehab-related content in fitness apps.

a 2021 study published by pubmed central cited that, “a global rise in the incidence of noncommunicable chronic diseases will cause an associated rise in the prevalence of disability and will be responsible for 75% of all deaths by 2030, thereby creating the most significant public health problem of the 21st century.”

yet large platforms like MyFitnessPal offer no guidance for users with temporary or permanent limitations, leaving them to research independently.

01

overview

The Main problem:

mobility limitations require
users to research and plan their own modifications,

ultimately leading to users feeling fatigued and frustrated.

Three separate images related to physiotherapy and mobility exercises. The first features a man with a caption about benefits of physiotherapy, including improving mobility, faster recovery, and strength. The second shows a person doing a leg exercise under the text 'High Hip Mobility Moves.' The third depicts a man in a squat position with text about fixing stiff ankles and mobilization with movement.
Three images related to physical therapy and stretching. The first shows a man in a therapy session with a group of people, with text about not stretching limitations that don't respond to stretching. The second features a man holding a sign that says "Stretching Isn't Mobility." The third depicts a man sitting in a rehab space with text advising not to skip steps in rehab.

project goals:

  • Define what makes an alternative exercise "safe" and "successful"

  • Understand how users modify and find exercises

  • Understand the effects of skipping or abandoning an exercise

  • Determine where user fatigue starts while searching for modifications


Research Questions:

  • What do users experience when searching for an alternative exercise?

  • What factors lead to an exercise needing modification?

  • What effect does seeing an alternate option have on users?

  • What makes some workouts feel successful compared to others?

02

research
& discovery

Methods:

  • User Interviews

  • Affinity Map

  • SWOT Analysis

Research participants:

  • users with temporary or permanent
    mobility limitations

  • used myfitnesspal

  • tracked workouts on a mobile app

Key user values:

  • Data Interpretation is the Greatest Asset

  • Responsive Feedback Sustains Motivation

  • Recovery Indicators Signal Readiness

A detailed infographic with sections, charts, and color-coded sticky notes discussing data interpretation, emotional reactions, and problem-solving strategies, including a map key and pattern breakdown on the upper right.

(Above) affinity map, patterns, and potential intercepting points for solutions

KEY FINDING

Users don’t want more data.

they need adaptable progress metrics
and interpreted data that aligns to their recovery goals and helps them feel acknowledged.

A woman with wet hair smiling indoors near a large window with white blinds, wearing a blue top.

“they (fitness apps)
just assume you’re healthy and have no problems.”

- karissa j.

refocus primary needs and solution

research shifted the original solution from:

How might we equip mobility-impaired users with modified exercises when they encounter movements they cannot perform?

to a user-informed solution:

How might we provide mobility-impaired users on MyFitnessPal with movements attentive to their current ability, responsive feedback, and actionable data that supports their progress and recovery goals?

03

user personas
& prioritization

personas

Two key personas emerged from the user data, differentiated between temporary and permanent limitations.

  • The resilient returner (short-term)

  • The Determined INnovator (long-term)


both resonated with the feelings of being dismissed by fitness apps while adjusting for or managing pain.

Methods:

  • MosCow prioritization

  • empathy maps

  • user scenarios

  • project v. business venn diagram

A woman in workout attire laughing and holding a smartphone in a brightly lit room, with a text overlay about fitness goals and challenges.
A woman lifting dumbbells while working out at a gym, with an overlayed fitness profile and motivational quote.

(left), The resilient Returner,
(right) The determined innovator

Goal overlap

Business goals

MyFitnessPal was a market leader in nutrition & strength-training tracking, however it has recently faced challenges related to:

  • maintaining user retention

  • establishing competitive differentiation

  • cultivating brand trust

Business needs

  • user onboarding

  • competitive edge (whoop, apple fitness, nike, etc.)

  • better brand approach, enhance the “my” in myfitnesspal

User goals

Users are adaptive, subject to change, and deserve tools that provide flexibility and motivation when these circumstances occur.

  • reduce cognitive overload

  • maintain motivation

  • feel confident to perform safe exercises

User needs

  • acknowledgement of their ‘start’

  • interpreted data

  • action indicators

  • pain-management feedback

  • body & intensity specific modifications

“interpreting the data gives the same value as paying a $175/month studio style membership.”

— Anna G.

features

A screenshot of a document listing features categorized under 'Must Haves' and 'Nice to Haves'. The 'Must Haves' section includes real-time AI-powered modifications and body-specific progress metrics. The 'Nice to Haves' section features Rate of Perceived Exertion (RPE) feedback.

04

structuring solutions

summary

Although the user flow was met with positive reviews for the resilient returner persona - lo-fi frames revealed an awkward and disruptive experience.

too much focus was placed on adjusting for impact and goal setting. emphasis doesn’t mean it needs to necessarily come first in the flow.

when transitioning into hi-fi, it became very clear the lo-fi frames needed to be simplified to match myfitnesspal’s flatter ui style.

“I'd want to get right to it and not wait until the end to put in my injuries..”

— Thomas F.

v.01 flexmode, onboarding prototype

05

Testing
& final impacts

summary

the v.01 hi-fi prototype was tested on a pool of 5 participants.

  • 30-59yrs

  • experience general limited mobility/injury/post-surgery recovery/chronic pain/arthritis

  • experience workout-related pain dure to condition(s) 1-4x a week

Users were evaluated on whether flexmode helped them feel more empowered or discouraged (key user need), and whether they shifted them from maladaptive to adaptive behaviors (feature adoption).

Afterwards, a survey measured scalable impact between users’ experiences with and without FlexMode.

Comparison chart showing three levels of success: high, moderate, and low, with usability and behavioral characteristics for each level.

(above) success metrics for usability testing

key useR refinements

  • User interface screen titled 'Misleading Selection UI' with options for Hips/Legs and Knees symptoms, showing selected severity levels: Moderate for Hips/Legs and Severe for Knees.

    Misleading UI Indicators - Checkmarks confused users who thought it implied multiple selections, changed to radio buttons.

  • Section titled '02 Generic Icons' with options for Upper/Lower Back, Hips/Legs, Knees, and Ankles/Feet, each with an icon of a plus sign inside a circle

    Generic Icons - Each body part had a custom icon to highlight the affected areas.

  • Screenshot of a user interface titled 'Redundant UI' with a section labeled 'Knees'. It lists two conditions, 'Arthritis' and 'Limited Mobility', each with 'edit' and 'remove' options, and both are marked with 'Severe' impact.

    Redundant UI - Convoluted composition and actions. Users felt it was confusing. Why take them back to different frames when they could return to one? (What’s affecting These Areas, V.01.4)

  • Section titled 'User Familiarities & Details' asking how today's workout felt, with options 'Good!', 'Okay', and 'Not good,' and a 'Rate Discomfort' scale from 1 to 5, with 4 selected.

    User Familiarities - Noted US users are more accustomed to seeing Negative → Positive order, “Rate Discomfort” needed defining metrics (“1=No Discomfort, 5 = Very Bad…”

v.02 flexmode, onboarding prototype

A digital graphic with a round photo of a smiling woman with light brown hair in the top left corner. The background is dark blue with a quote in white and light blue text: "FlexMode would be nice. I'm still hitting the goals I want to do but it's also adjusting in a way where it takes into consideration what I need."

reflections

I realized too late that MyFitnessPal was pretty outdated compared to when I had used it in the early 2010s.

I was concerned it was an insufficient platform to test on since it was difficult to reach familiar users. However even with an older, flat platform I was still able to find an impactful solution for modern users and gained a new outlook on identifying critical user gaps.

A quote from Chris L. on a blue background emphasizes the importance of receiving immediate feedback for rheumatoid arthritis symptom patterns, with a small circular profile picture of a person with a dark beard and short hair against a gray and blue background.
Quote from Megan O. praising a feature for being hard to maintain streaks due to injuries, displayed on a blue background with a user icon at the top.
Comparison chart showing the impact of using Flexmode on exercise modification and pain continuation. Without Flexmode, 20% modify exercises, 1/5 modify to safer exercises, and 2/5 continue despite pain. With Flexmode, 100% modify exercises, 5/5 modify to safer exercises, and 0/5 continue despite pain.

When one of the users told me, “They just assume you’re healthy and have no problems,” it made me realize a lot of apps probably do this and unintentionally ignore some major user needs in the hustle of plotting a solution. That new perspective gave me a launch pad to examine where else users have fallen through due to major gaps like this.

The boundaries provided great learning tools in how to focus creative solutions within established brand guidelines and how to solve for the most impactful user need without the expansive freedom of starting from scratch.

however, I was too granular around adaptability metrics and what that looks like, that I didn’t think of how this would integrate into their current goals aside from their preexisting workouts/calorie counts.

if i were to redo this process, i’d be more mindful to flex in between high-level flows and pixelated details.