Electric Motorcycle Rental — Automated Rider eKYC
How a Southeast Asian mobility startup replaced 3-day manual KYC with a 15-minute AI-driven onboarding flow — verifying identity, income, and gig earnings from Grab and Gojek screenshots.
Riders applying for electric motorcycle rentals to work on Grab or Gojek faced a 3-day onboarding bottleneck: video KYC calls, manual document review, and ad-hoc income verification from app screenshots.
The Challenge
Manual verification couldn’t scale with the fleet’s growth. Every applicant meant a compliance reviewer cross-referencing a government ID, a selfie, and inconsistent income proof — earnings screenshots from ride-hailing apps rarely matched any template.
The Solution
An AI-driven eKYC flow replaced the entire funnel:
- Face match + liveness check against the uploaded ID
- OCR extraction from Grab and Gojek earnings screenshots
- Cross-validation of weekly ride counts against declared income
- Confidence-scored auto-approval with escalation only for edge cases
Business Value
Manual video KYC dropped by 75%. Onboarding time went from 3 days to 15 minutes, and the fleet now onboards 40% more riders per month with the same compliance headcount. Fraud detection improved 3× — AI catches forged earnings screenshots and document inconsistencies that reviewers frequently missed.
Wrapping Up
Use-case pattern: when verification data lives across multiple user-generated document types, combining OCR with cross-field validation outperforms any single check. The approval loop keeps humans in charge of only the low-confidence tail.