Case study

EasyPark / ArriveDiscount flow: from 80% abandonment to 90% conversion

Mobility · Fintech

Redesigned EasyPark's discount application flow on 457,000 annual transactions with 80% abandonment. Two interventions: inverted the input hierarchy (text first, not camera scanner) with auto-apply — then added a proactive checkout prompt and embedded apply actions per card. iOS: 20% to 90% conversion, 32 seconds to 5. Android: task time improved 37%, conversion barely moved — an open problem, not a complete win.

Context

EasyPark runs ParkMobile in the US under the Arrive Group. One flow processes 457,000 discounted transactions per year. 80% abandonment rate. The problem wasn't that users lacked discounts — the interface actively worked against them.

Challenge

Usage data showed 99% of users entered codes manually. Yet the interface led with a camera scanner — every user had to tap past it to reach a keyboard. Conversion was also failing upstream: users reached checkout and forgot they had discounts entirely.

My role

Senior product designer on the discount flow. I owned the usage data analysis, decision framing, and both intervention designs.

What I changed

  • Inverted the hierarchy: text input became primary, camera scanner secondary. Added auto-apply — enter a valid code, it applies immediately, no second confirmation step. Task time: 35s→24s Android, 32s→5s iOS.
  • Added a dynamic checkout prompt ('2 discounts available') to surface discounts at the point of highest intent. Replaced the disconnected Apply button with embedded Apply actions inside each card — selection and confirmation became one gesture.
  • Mapped all four backend error types to specific states with specific recovery actions. Generic 'invalid code' replaced with precise failure messaging.

Outcome

iOS: purchase conversion 20% to 90%, task time 32s to 5s. Android: task time improved 37%, conversion stuck at 22%. Same architecture, same UI, different result. Bottleneck shifted from friction to engagement — open problem at end of engagement.