Interview Questions › Root Cause Analysis

Signup-to-KYC rate dropped after UI revamp

Following a modern UI redesign of the identity verification flow, conversion from initial registration to completed KYC verification dropped from 68% to 51%. Diagnose the UX and technical failures.

Context

The revamp intended to reduce cognitive load by splitting the document upload into a 4-step progressive modal.

Clarifying questions to ask first

• At which specific screen in the 4-step flow are users dropping off?
• Did document verification rejection rates increase or did users simply not upload?

How to approach it

1. Analyze the Step-by-Step KYC Funnel: Personal Details -> ID Document Selection -> Document Photo Capture -> Liveness Selfie -> Automated OCR / AML Verification -> Instant Approval
2. Examine OCR recognition rejection rates and camera permission drop-offs on mobile browsers
3. Evaluate UI design changes (clarity of instructions, glare warnings, manual retry messaging)
4. Segment by Document Type (Passport vs Driver's License vs National ID Card) and Device OS

Framework: Step-by-Step KYC Funnel Isolation -> Document Photo Capture & OCR Latency -> Camera Hardware Permissions -> User Guidance & Error States

Sample answer

• Camera webview auto-capture triggering blur and flash glare rejections
• Removal of explicit guidelines on acceptable national ID types
• Mobile camera permission modal failing to reopen after accidental denial

1. Step 3 (Document Capture) drop-off doubled: new real-time auto-capture had an overly strict edge-detection threshold, rejecting 44% of valid ID photos with an unhelpful generic error

Overly sensitive camera auto-capture threshold blocked valid IDs; added manual capture fallback button, instant glare guidance, and relaxed blur tolerances to recover 17% conversion.

What interviewers look for

• Signup-to-KYC Completion %
• OCR Extraction Success Rate
• Camera Permission Grant %
• Document Re-upload Rate