Confidently Onboard Good Customers

An identity document serves as proof that people are who they say they are. Generally, the documents are government-issued, unique and include personally identifiable information.

The guidelines for acceptable identity documents vary by use case and country. Driver’s licenses, passports or national identity cards are common identity documents.

Enabling people to sign up for services on an app or website widens the potential market and makes commerce easier for customers. Automated digital identity verification can decrease manual work and improve productivity.

Document verification typically starts with a person taking a selfie and a picture of a driver’s license, passport, national ID card or other identity document. The images are then checked for evidence of tampering and to determine if the selfie matches the document portrait.

Machine learning technology detects and automatically captures the document and selfie images to provide a smooth customer experience, optimize image quality and achieve superior verification rates. Machine learning is also critical in classifying documents and detecting fraud by comparing selfie and document images.

Verification rates play a significant role in determining the number of customers a company onboards. Poor document image capture can lead to increased manual reviews and slower onboarding, so a guided user experience can be critical to document verification success.

The U.S. National Institute of Standards and Technology Face Analysis Technology Evaluation ranks the performance of biometric algorithms. The Trulioo KYC Documents biometric algorithms rank in the top five out of 82 for security (maximum fraud attempts detected) and convenience (minimum false positives).