Buy Now Pay Later

SmartMoney is a buy now pay later service that allows users to buy items now and pay later, or use installments to pay for them over time. They operate on pretty tight margins, because a portion of their microloans end up defaulting.

SmartMoney is also wary of raising fees or interest rates, because this is a very competitive market and they're trying to scale to capture a significant market share. They tried utilizing background checks to identify high risk customers, but this was very expensive and they found a lot of their customers are high risk - in fact catering to these customers is a big part of their business model.

As they've taken control of their spend, they find they're on a good road to profitability and their business model is sound. However, they're facing a new threat that's very rapidly eating into their margins: fast paced fraud.

Fraudsters open accounts with fake documents, use the accounts to buy items, and then default on their payments. Usually, they'd be able to legally pursue the lendees or at least sell the loans off, but that's obviously not possible for these cases. The main issue with this type of fraud is the sheer scalability. The fraudsters are able to open accounts at a rate that's far greater than the rate at which they can be detected through manual means. By the time they're identified and SmartMoney can put in counter measures, a lot of damage has already been done.

SmartMoney implemented counter measures like escrow systems and delayed payments, but they find these create too much friction, and they're losing out on new customers.

Location	Fraud Risk	Fraud cost	False Rejection Cost	End user motivation
Remote	High	High	Medium	High

SmartMoney are looking for a fully automated High Assurance solution, because each instance of fraud costs them a lot of money directly. They can afford some friction if Verify can solve all their problems, because their current mitigations are very high friction in comparison.

Here's how they could configure Verify for their needs:

{
    "UseCase": {
        "VerificationPolicy": "Strict",
        "VerificationContext": "Remote",
        "ManualReviewStrategy": "ReviewRejectedOnly",
        "ManualReviewSensitivity": "Default"
    }
}

This setup works really well for them, and the hit to their real customers is negligible. There are some false rejections (significantly less than 1%), but even most of those end up in ManualReview, and they're able to recover most of those customers. Rarely, some fraud attempts do succeed (around 1-2% of all fraud attempts), but they're very hard to scale. Fraudsters can't use the same documents multiple times, so even when they succeed once in a blue moon, it's extremely hard to do it again. Soon enough, most of them figure out it's not worth the effort and move on.