When it comes to satisfying the consumer need for instant gratification, the internet has disrupted a number of industries. Take, for instance, online lending. RocketLoans, a subsidiary of Rocket Holdings (the parent of Quicken Loans), offers loan approvals in as little as 10 minutes with funding in under 24 hours. That might sound remarkable, but it’s the new normal. Several other companies offer a similarly rapid approval process.
What’s perhaps even more remarkable is that these online lenders aren’t just targeting traditionally “safe” borrowers, those with long established credit histories and good credit scores. They’re also promising the same speedy approval process to the more than 64 million “thin file” customers in the U.S., people who have little or no credit history and often aren’t served by traditional banks and lenders. It’s a huge potential opportunity in a market that’s projected to reach $350 billion by 2020, but one that brings additional risk of fraud.
For online lenders, managing that risk is a challenge in what has become, essentially, a real-time marketplace. While they might want to take more time to review an application and verify the identity of the person behind it, lenders don’t have that luxury. A survey by PwC, found that one in three borrowers not only place an emphasis on the speed of the application and approval process when choosing a lender but that they value it even more than a lower interest APR. The takeaway for lenders competing for their business is that borrowers have choices and they’re not afraid to take their business elsewhere if they can get money faster.
But how can lenders meet the demand for rapid approvals without shortchanging the identity verification process? First, it’s important to understand how online lenders come by their customers. While some capture leads directly through their own online storefronts or portals, many others purchase leads from online lending marketplaces (sometimes called ping tree marketplaces). The ping tree provides a centralized marketplace for multiple companies ranging from lead generators and wholesalers to retail recipients, matching providers and buyers based on transaction type, field requirements, and pricing – all within milliseconds.
When they evaluate leads for purchase, most lenders perform a few basic checks. Is the borrower from a geography where the lender is authorized to do business? Has the borrower been applying for multiple loans simultaneously or in quick succession (a fraud tactic known as loan stacking)? Is the borrower on any lists of known bad actors? Lenders also check with alternative credit bureaus that can see whether a borrower has a history of staying current with bills from cell phone companies, utilities and other service providers. While not a credit history per se, on time bill payment can be indicative of whether a borrower is likely to pay back a loan.
While these simple checks are important, they are essentially binary. They can’t really provide context about the borrower’s true identity or a meaningful indication of the quality of the lead. Without evaluating the veracity of the metadata provided by the applicant, it’s hard for lenders to know whether there’s a real, contactable customer behind the transaction.
It’s here, at the top of the funnel, where non-personally identifiable information (non-PII) data can make a difference. Non-PII data (and the linkages between data elements) gives lenders a much more powerful indicator of the quality of leads. Comparing the non-PII signals from prospective borrowers to those of their best performing customers can give lenders a more informed perspective on the quality of leads they acquire. Good leads can be fast tracked for approval while more questionable leads can be flagged and either rejected outright or sent for further review.
Positive signals include things like:
- email address age (more than 720 days is good)
- IP address proximity to physical address (within 10 miles is good)
- phone and address match
- email and name match
- phone and name match
- address and name match
Some risk signals include:
- linked email, phone or address details that don’t match
- an email address that is less than 90 days old
- a non-fixed VoIP or toll-free phone number
- the phone country code and physical address don’t match
- phone, email or address are invalid
- a proxy IP address
Strong positive non-PII data signals are a good indicator that a borrower is a real person, and not a manufactured identity. Because fraudsters are continuously evolving their tactics and coming up with new exploits like synthetic identity theft (combining real and fake identity details to fool fraud systems), lenders need to rely more on data that is continuously updated and analyzed.
For lenders, being able to quickly distinguish between a real customer and a fraudster not only speeds up the transaction, but also reduces the chance that a good customer is made to wait or, worse, has their transaction rejected (also known as customer insult). Customer insult is no small problem; not only is there an immediate loss of revenue, but a rejected customer is unlikely to return in the future. On the flip side, customers that enjoy a speedy transaction are much more likely to recommend an institution to friends and family.
Non-PII data (and the linkages between data elements) applied at the top of the funnel ensures that only the best leads make it into the funnel. It can also help at the bottom of the funnel. With better qualified leads, lenders can see better take rates on loan offers. And with the reduced risk of fraud comes lower First Payment Default rates.
Of course, none of the potential benefits of non-PII data would matter if using it slowed down the transaction process. Fortunately, the data can be integrated into most fraud management systems without introducing noticeable lag. That means lenders can continue to offer speed while improving identity verification.
Tom Donlea leads the global marketing efforts of Whitepages Pro, the worldwide identity verification data provider for risk management in banking and online lending. With over ten years of online payments and risk experience, he previously was the founding executive director of the Merchant Risk Council.