It might seem crazy to offer loans to people from all over the world with only their Facebook profile as reference, but Lenddo has created an algorithm that makes this not only possible, but apparently successful.
Lenddo is a mobile app that allows users to build a profile and connect it to their social media accounts. It then uses data from those accounts to assess the credit risk of candidates applying for loans.
The concept came from a clear gap in the market: the burgeoning middle class, for whom low credit scores can make it difficult to be eligible for loans and make the economic jump forward. Lenddo’s Founder and CEO Jeff Stewart estimates that this group includes 1.2 billion people worldwide.
The app not only uses data from social media, but utilizes the network of friends users cultivate online. In order to create accountability among users, those applying for a loan can select a group of Facebook friends to use as references to help them qualify. However, if the loans are not repaid, the friends also become less creditworthy.
The concept of using group accountability is familiar to traditional community-based micro-finance lenders.
In order to prove that social media could serve as an indicator for credit worthiness, Lenddo founder Jeff Stewart conducted a three-year test. The product’s first market was the Philippines, but since then loans have been distributed all around the world including Colombia, Mexico, Indonesia and parts of Africa.
So far Lenddo’s alternative application process has lead to 10,000 loans and 500,000 members worldwide. The interest rates average 2–4 percent per month, comparable to those offered by other financial institutions.
In the Philippines, where Lenddo is most fully integrated, there are currently 250,000 users, enough to make the check and balances system of holding Facebook friends accountable for their loans has proven successful.
“We had to test out all the parts of our thesis – including making loans to people we knew wouldn’t pay us back – to make sure it worked,” Stewart told WIRED. “By last summer, we finally had enough data to show that the score worked.”
Now that Stewart knows his algorithm works, Lenddo will no longer be lending. Instead it will sell the algorithm to other financial institutions.
The logic behind Lenddo might seem questionable, but they are far from the only startup working off this model. Cignify’s product judges loan risk by analyzing cell phone activity and Affirm uses social data to increase credit scores for millennials. But at this point, Lenddo has an edge in terms of research.
Brett King, CEO of money management startup Moven, thinks that Lenddo’s method is less risky than the traditional loan application process.
“When you have people fill out an application form, you only get that data plus a credit score,” King told WIRED. “Today, we can bring so much more data to that decision that now an application is taking on more risk than you need.”
Studies, including one recently published by Stanford University, have shown that behavior on social networks has a strong correlation to personality and serves as a rich source of data on a person’s character. Facebook may know your personality profile better than even your close family members.
“The evidence increasingly shows that [data mining] can say a lot about a person,” says Michal Kosinski, a Stanford University professor who co-authored the study. “Our most recent study shows that our computer can beat your close relative in determining your personality. But computers can also predict your sexual orientation, your political views, and so on.”
From its conception, Lenddo founder has said that the digital profiles of loan candidates offer better evidence about whether or not they can repay a loan than a credit score does. While Lenddo’s algorithms are unlikely to replace the traditional credit score vetting system any time soon, Stewart sounds confident that the company will be powerful player in the financial sector.
"We don't see why this can't be one of the most important companies, like, ever," said Stewart.