pounce
Fitness apps lose beginners once tracking turns into a chore. I led a team of four designers through a 7-week sprint to design Pounce, a fitness app built around friends and a cat mascot. After our fixes, task completion rose 35% in the second round of testing.
4 min read
overview
the problem
People don’t quit fitness apps for lack of features. They quit because the apps feel like work. Our 12 interviewees (ages 18–25) wanted a supportive companion, not a productivity tool. Strava and MyFitnessPal have strong goals and communities, but felt demanding to beginners.
So the brief became: keep it simple, make progress visible, make it feel personal. A feature matrix cut everything that didn’t fit seven weeks, leaving home, crews, statistics and profile.
key decisions
give the mascot a job
finding — every tester finished both tasks and loved Pounce, but many couldn’t say what the mascot was for.
"The concept is really cute, but as a first-time user I wasn’t sure where to begin or how Pounce connected to the experience."
— usability test participant
Pounce levels up with you — logging moves it forward, so there’s a reason to come back.
Crews share a mascot — each crew’s cat has a level and an XP bar (e.g. “Lvl 10 Puffytron, 100/150 XP”) and a mood line: “Puffytron is feeling super!”
Group goals, not leaderboards — a crew works toward one shared weekly goal, with group progress instead of a ranking.
trade-off — a levelling system is more to design and explain than a cute character, but a mascot with no purpose was decoration.
stats first, actions second
finding — testers struggled to find their own progress on home.
This week, at a glance — steps, calories and heart rate as cards right under today’s goal ring.
One primary action — a single “Log a Workout” bar replaced two competing cards.
Fewer steps — the most-used features were cut down to fewer taps. Task completion rose 35% in round two.
trade-off — the stats took the space the challenge cards had: clarity over novelty.
onboarding one question at a time
finding — half of our survey respondents had quit an app for being too complicated.
A greeting, not a form — “Hi, I’m Pounce. Let’s get your paws dirty…” before any question.
One choice per screen — name, sex and birth date (to estimate daily energy), activity level (four, each defined), main goal (six tiles), days per week.
Commitment you can keep — 3, 5 or 7 days, shown as weekday chips. 3 is labelled “Great for beginners”.
Always skippable — a progress bar, Back and Skip on every step; it ends with “You’re all set”.
what testing changed
visual system
Bright, not clinical — pink, blue and yellow on white. Interviewees were tired of dark, cluttered fitness apps.
Rounded everything — buttons, cards and icons in soft shapes that feel welcoming to new users.
One typeface — DM Sans, modern with a little personality.
constraints & reflection
Seven weeks, one designer down — one of five designers left late in the sprint; we split their pages and still delivered.
A prototype, not a product — the 35% is from two rounds of moderated testing, not live use.
Judges’ challenge — Ken Skistimas (ServiceNow) asked what Pounce would look like if its AI ran through the whole app.
Next: map Pounce’s growth to real milestones (streaks, finished crew challenges), let the AI write encouragement on home instead of only in chat, and add watch and home-screen widgets for logging at a glance. Testers loved the mascot before they understood it; next time I’d test what a character is forbefore we polish how it looks.