SPEC CASE STUDY

gymcity - outdoor workout iphone app

A fitness app allowing users to work out in outdoor public spaces using trainer-sourced workouts.

Role:

UX Designer
·

PLATFORM:

iOS

Duration:

4 weeks
·

TOOLS:

Sketch, Proto.io, Lookback

Problem

With the closure of gyms, the pandemic limited people’s fitness options. GymCity addresses the need for equipment-free workouts, and for doing them outside as users are increasingly confined to their homes.

Goal

These were the guiding objectives before any user research.

  • Offer trainer-led, equipment-free workouts in nearby outdoor spots.
  • Build social accountability through challenges and friend updates.
  • Let busy users filter by time, location, and workout type in seconds.

Research insights

5 interviews with gym-goers turned home exercisers revealed:

  • Expert guidance is non-negotiable.
    YouTube “follow-alongs” beat ad-hoc routines.
  • Accountability > camaraderie
    Four of five value motivation more than socialising.
  • Most workouts already happen outside or need minimal gear.
  • Time drives adoption
    quick access, short sessions, progress tracking keep people on plan.Four of five value motivation more than socialising.

Design

I aimed for a map-first mobile flow that shows nearby workouts, social nudges, and quick progress checks without heavy onboarding.

Early flow mapping ensured GymCity covered the five needs surfaced in research:

  • Social accountability
  • Community interaction
  • Progress tracking
  • Challenges
  • Trainer transparency

Key Screens

Home (map view)

Browse trainer-authored workouts by location; filter by time and intensity.

Workout details

Clear steps, minimal gear list, start button triggers session timer.

Challenge hub

Join challenges to gamify progress and boost accountability.

History & stats

Glanceable streaks and personal bests keep momentum high.

Design Trade-offs
  • Map-first browse vs list view
    Map chosen so location benefit is visible at launch.
  • Friend updates vs full history feed
    ‍    Opted for push updates only; private logs prevent data-sharing concerns.

Testing

Five remote Lookback sessions covered onboarding, filtering, workout start, challenge join, and stats lookup.
Key fixes:

  • Moved filters to Home only (onboarding no longer points to Profile).
  • Limited friend visibility to single workout alerts; removed full history feed.
  • Adjusted swipe regions in the prototype to avoid gesture traps.

All five users then completed every task without confusion.

Lessons

  • A smart algorithm still needs human context.
    Whenever the “cost of a bad pick” feels high, users lean on people they trust, even if the platform’s recommender is world-class.
  • Prototype small, test early, refine fast.
    Even lightweight studies (5–6 users) can uncover blind spots—like unclear audience labeling or risky bulk-send defaults—long before code is written.