All work
05Mobile app

Guest Pass Mobile App

QR check-ins and conversion analytics that digitise gym guest entry.

Role
React Native development, scanning and analytics integration
Type
Mobile app
Stack
5 technologies
Live
yourguestpass.com
guest-pass-mobile.app1/3
Check-in experience in the app.
Check-in experience in the app.

The brief

What needed solving

Guest entry at a gym is still paper and front-desk judgement: a visitor arrives, someone writes a name down, and whether that visitor ever becomes a member is nobody's data. The app had to make entry take seconds at the door, while turning each visit into a signal the gym owner can actually act on.

Approach

How it was built

01The door is a five-second interaction

Everything about the check-in flow is shaped by where it happens: a queue at a front desk. The scanner opens straight to camera, results are unambiguous at a glance, and both the approved and rejected states are legible from arm's length without reading a paragraph.

02Verification that survives real conditions

Scanning happens under bad lighting, on cracked screens, at odd angles. Recognition was tuned for those conditions with a manual code-entry fallback always one tap away, so a failed scan never becomes a blocked entrance.

03Visits as data, not paperwork

Each check-in feeds behavioural analytics that model how likely a guest is to convert to membership, giving owners a ranked view of who to follow up with — the same action that previously produced a line on a clipboard now produces a decision.

Scope

What shipped

  • QR-based guest check-in with instant approval or rejection
  • Computer-vision-assisted verification at the door
  • Manual code fallback when scanning is not possible
  • Membership conversion probability scoring for owners
  • Visit history and behavioural analytics per guest
  • Location-aware gym lookup via Google Maps

Engineering notes

The hard parts

Challenge

Camera and vision work is the heaviest thing a phone can do, and it has to stay responsive on the low-end Android devices that front desks actually use.

Solution

Processing is throttled to sampled frames rather than every frame, with the camera surface released the moment a scan completes, keeping battery and thermal cost proportional to a short interaction.

Challenge

Gym networks are unreliable exactly where the scanner lives — thick walls, weak signal, dead spots at the entrance.

Solution

Check-ins are queued locally and reconciled when connectivity returns, so a lost signal delays reporting rather than stopping people getting through the door.

Challenge

Predictive scores are worthless to an owner if they can't tell what to do with them.

Solution

Scores are surfaced as ranked, actionable follow-up lists tied to individual guests and visits, rather than as an abstract percentage on a dashboard.

Outcome

Where it landed

  • 01Paper guest logs replaced by a seconds-long digital check-in
  • 02Every visit converted into structured data owners can act on
  • 03Entry keeps working through weak connectivity and failed scans

Have something like this to build?

This project went from design through deployment. Yours could be next.