3 ways automated document verification software speeds up onboardingMarch 1, 2023
By automating the verification of important documents like IDs, passports, and licenses, businesses can expedite the process, reduce manual effort, improve accuracy, and aid compliance. In this article, we’ll explore three key ways that automated document verification can help organizations onboard new customers or employees more quickly and efficiently.
Who are you?
In our modern communication world, that’s a question typically answered with an identity document (ID). For highly-regulated and fraud-vulnerable products and services, ID verification is a practical, often seamless way to confirm a user’s identity.
Digital banks, payment apps, home delivery services, and online gambling sites are just a few examples of where some level of ID verification is required, and up until recently, the process was executed inefficiently. Either by including a human-in-the-loop or through secondary applications and services, identity verification was marred by slow processing times, false positives, and other roadblocks.
From the consumer’s point of view, that’s frustrating. One survey found that 26% of financial services customers who dropped from onboarding did so due to experiencing friction in the identity verification process.
Fortunately, smoothing out that friction and reducing lost revenue is possible with the right solution.
Let’s talk about how it all works, and how automated identity document scanning and extraction can ensure user onboarding is fast, easy, and accurate.
What is automated identity document verification?
From a user’s perspective, automated identity document verification involves scanning the front and back of an ID using the camera on their mobile phone. This can capture and extract data from the document in mere seconds. From there, the relevant PII (personally identifiable information) can be automatically populated into a field.
This is ideal for consumers signing up for online bank accounts or other info-laden registration processes. It’s also ideal for organizations looking to streamline their data capture processes without sacrificing accuracy, as was the case with Omnicare Medical Center, a healthcare provider that needed to efficiently and accurately enter new patient info into its COVID testing database.
Beyond the scanning and data extraction, however, is the actual process of identity card verification. An ID verification solution needs to be robust, in order to truly determine the legitimacy of the document in question. That’s where technology (especially the growing field of AI) comes into the picture — quite literally.
Here are three ways automated, AI-powered ID verification is reshaping the industry.
Comparing ID fields for consistency
Nowadays, almost every ID has the document holder’s personal data encoded in some way, whether in a barcode or a Machine-Readable Zone (MRZ). Any mismatch between this data and the data printed on the rest of the document (i.e., the Visual Inspection Zone or VIZ) is a telltale sign of tampering.
Finding a solution that can distinguish and interpret the different zones on an identity document is difficult enough. The problem beyond that is IDs vary wildly from country to country.
Reading unstructured data requires deep document expertise and flexibility on the part of the ID scanner. The software needs to know that on a Malaysian ID card, for example, the first six digits of the document number signify the person’s date of birth and that the corresponding digits are based on their place of birth and gender.
That’s where AI-based solutions really stand out compared to less intelligent methods. Thanks to extensive training using machine learning, a sophisticated ID scanner is capable of better understanding the context of the information it’s reading and extracting. Within seconds, the solution is able to verify that there is consistency across all ID fields, including comparing the back of the document to the front.
Checking on security features
Every ID comes with its own unique set of security features, including holograms, logos, watermarks, and microprint. Checking for the presence and location of these security features is a crucial step in verifying if a document is genuine or fraudulent.
Complicating the authentication of these diverse feature sets, people tend to take ID photos that are blurry, obscured, low in resolution, or held at an unreadable angle during scanning. In fact, studies have found that images of insufficient quality were the main reason for failed verification of identity documents.
As a result, one of the keys to effective ID verification involves snapping the clearest possible image of a document, to remove human error or camera quality from the equation. In this instance, all the user has to do is point and click.
Intelligent document verification software does that automatically, by first detecting the document, checking for any blurring or other hindrances, and then separating it from its background. The end result is a clear, horizontally aligned image of the document for the software to verify.
Matching the user’s face and ID portrait
Just because an ID card matches the face and presumed info of the user, how do we know the document is authentic? There are also instances where fraudsters create synthetic IDs, which combine fake identities with real PII aspects (e.g., social security number).
One way organizations try to combat this is by asking a user to take a selfie and then comparing it to the photo on their identity document.
In this instance, sophisticated technology, trained on diverse ID types and real world images, is basically trying to recreate – and automate – what a bank teller, cashier, or TSA agent does when verifying customers in person.
Streamlining this step hinges upon the ability of the core technology to capture a clear portrait of the user before comparing it to their selfie. If the distinctive physiological characteristics on these two images match, there is a high likelihood the user is a genuine owner of the submitted document.
The bottom line: Streamline onboarding with automated document verification
From the perspective of a company looking to onboard as many happy customers as efficiently as possible, automated online document verification is a key step towards improving user and business outcomes.
Today’s users expect more — or in this case, less — from their digital interactions. They don’t want to spend hours (sometimes days) waiting on verification results. Rather, they want to get through onboarding and authenticating processes fast and easy, with ID verification being just a mandatory step in the way.
And with rapid technological improvements driven by AI/ML, organizations can do this with confidence. With intelligent ID scanning and extraction software, you can feel positive that customer information is correct, without tasking employees with the tedious activity of doing so.
For any organization still lagging in this department, it’s high time to make the ID verification experience seamless — improving your customer experience in the process.