Verifying someone’s identity– making sure they are who they say they are– is a key facet of fraud deterrence. Without it, it’s very easy for bad actors to assume someone else’s identity to steal money and/or information from businesses and unsuspecting individuals.
There are four main ways to verify someone’s identity with document identity verification:
- Having an employee check the person’s government-issued ID in person
- Leveraging an ID verification software solution to instantly and automatically check government-issued IDs
- Having a customer submit an image of their government-issued ID and selfie online for an employee to remotely verify
- Using security questions and two-factor authentication to verify a customer’s identity.
All of these methods enable businesses to verify someone’s identity, but some are more accurate, cost-effective, and faster than others.
Manual ID Verification
In general, manual document ID verification refers to an employee checking a customer’s government-issued ID by hand– whether in person or over the internet. While human error is certainly a risk when it comes to manual verification, delegating employees to verify IDs can also impact processing times and impede growth. For many businesses, speed and accuracy can be a key differentiating factor for customers to choose to spend their money with you over someone else. When customers are left with a less-than-desirable ID verification experience– as is often the case when there’s another person at the end of the ID verification process– they’ll go elsewhere.
However, a manual ID verification solution does enable businesses to have added control over the ID verification process as an employee manually checks every single ID that goes through the system. And it can be an attractive starter option for newer businesses who don’t have the resources or customers to justify exploring alternate solutions.
Automated Document ID Verification
An automated document ID verification process typically refers to software solutions that verify government-issued IDs using machine learning or optical character recognition (OCR). These solutions verify someone’s government-issued ID by looking at the machine-readable aspects of an ID– the barcode, expiration date, birthday, and image– and leading solutions compare the ID’s image with a real-time selfie taken by the user, verify the ID isn’t expired, and check to make sure the person’s age is above the threshold set in the identity verification workflow. Verifying someone’s identity through an instant and automated solution is typically much faster and easier– many solutions allow businesses to set different identity verification flows by location to stay compliant across multiple cities. For example, a car rental company could set it so that drivers located in San Diego will only be able to rent a car if they’re over 25 years old while drivers located in Los Angeles can rent if they’re over 30. This negates the need for additional headcount or employee training, reducing the chance for mistakenly accepting an ID that shouldn’t have been accepted.
And because the process is so much faster and more accurate than manual verification, many businesses have found these solutions to catch more fake IDs and decrease the amount of time it takes customers to onboard.
Which method is best for you?
ID verification isn’t a “one-size-fits-all” situation because fraud doesn’t work that way. Depending on the sector, businesses will experience a variety of different types of fraud. For example, businesses that sell age-restricted goods like alcohol will, on average, most often deal with underage fraudsters who aren’t as sophisticated. Fintech companies, however, will most often deal with much more targeted bad actors who are very sophisticated and are willing to spend more time working to exploit vulnerabilities in their systems.
Regardless of if you choose to hire someone to check IDs or purchase a software solution to do so, make sure you’re choosing the method that will fit into your existing fraud deterrence plans, scale with you, and won’t put undue burden on your legitimate users.