During our first session of Fraud Office Hours, an attendee asked, "What’s an effective way to layer in frictionless fraud prevention at onboarding?" Watch this video clip and read below to see how Unit21's Head of Fraud Risk, Alex Faivusovich, responded.
Layering in Frictionless Fraud Prevention at Onboarding
"When it comes to onboarding, (and I think this refers to Fintech companies and crypto companies who, two years ago, were doing this), there was a mindset of doing 'onboarding at all costs. Let's bring volume.'
Like, 'let's bring as many people in as we can. Let's make huge campaigns in social media and try to open the top of the funnel as much as we can.' But I think now, especially with the stressful economic climate that we are entering, we're seeing a much more balanced approach when it comes to onboarding.
So frictionless onboarding can be achieved if you take all the signals that you get from the applicant during his onboarding journey. You look at them separately, so you understand all the different elements of the PII. You understand what risk is associated with their email, what risk is associated with their phone number, and what risk is associated with their address. How the correlation is working between the actual entity and the email, phone, and address.
What is the risk associated with the device? Can we learn something about the device? Do we think this device is a legit device? So maybe somebody is using some type of emulator. So if you can take all those elements apart, (and this is also a good way to look for an identity verification vendor), try to find a vendor who gives you all those different elements and give you insights on all those different elements.
Then, you can pretty much start layering your risk strategy around those different PIIs and say, 'Okay, if the PIIs look good, let's see how the email looks. Let's see how the form looks. Let's see what we know about the address.' And put different thresholds and different signals in place to ensure you're actually catching the bad guys."
Automate Onboarding Flows with Unit21
With Unit21, teams can create workflows for different potential customer journeys, allowing them to automate portions of the onboarding process so they can sign up customers quicker and with less friction. Workflows can be established that leverage rules to automatically approve or reject new users.
If teams see low scores for all risk signals for a user (such as synthetic ID risk, identify fraud risk, email risk, phone risk, or address risk) and the user doesn’t appear on any alert lists (such as sanctions), they can automatically accept the user. If a user exhibits high scores for any of the signals, they can automatically be rejected. If neither of these criteria are met, the user can be sent to secondary screening, where we can factor in other data points, such as trust or risk scores from third parties.
Keep in mind that Unit21 can ingest any data point from any third party providers, and then teams can use Unit21’s verification platform to write rules on this data that allow teams to automate a frictionless onboarding flow.
It’s also crucial for teams to monitor the performance of these systems, checking how many cases are automatically approved, rejected, or escalated to manual review. This is a great way to understand how balanced your strategy is. If teams see a spike in rejections, it’s a sign that either the process is too strict and you’re checking for too many signals, or you’re under some type of targeted attack. This empowers teams to identify their biggest threats, or shortcomings in their onboarding flow, and then improve for better performance.
Looking for more insights? Check out our first session of Fraud Office Hours on-demand for a deeper dive into current fraud trends and which preventative measures to consider.