The digital footprint: a tool to increase and improve lending
Analyzing someone’s solvency based on what they do on social networks is already boosting access to credit in emerging economies. In developed countries, it helps financial institutions better determine to whom they lend money.
Marta is a woman who is almost 50 years old and her bank did not approve a loan to remodel her kitchen. She believed this was because her checking account has never been very buoyant, but she has no idea that her comments on Facebook played a big role in the financial institution’s decision. Specifically, her emphasis on spirituality, the soul and the afterlife worked against her.
“We have data showing that people who write a lot about the soul have high levels of default. They are concerned by what will happen in thirty years, but not the fine print of everyday life.” This is how Pentaquark Consulting CEO Estela Luna, explained why the credit scoring model her company developed does not trust customers who are “too spiritual” at the Open Expo, an Open Data event recently held in Madrid.
Pentaquark Consulting has developed a system based on algorithms and machine learning, which studies social networks to determine bank customers’ solvency. It’s not the only company that monitors Facebook, Twitter, Linkedin, etc. to get more clues on social network users’ wallets.
The fact that a bank takes a look at what you put on Facebook may bother you, but it can actually work to your benefit. Luna says that one of her customers gave them the goal of “increasing the number of approved credit applications by 10%.” In other words, it’s possible that social networks show that they have more potential than the isolated financial data the bank has.“ Many people have a somewhat low income, but then we see on social networks that they go to the Riviera Maya Muchos on vacation.” And the benefits of using social networks for credit scoring are even greater in emerging economies.
The reason for this is because they are markets where a large percentage of the population does not have access to banking services, which automatically excludes them from traditional credit sources. Banks do not lend money to someone they know nothing about. However, technology is starting to break down this barrier.
The Singaporean company Lenddo and German company Kreditech, in which the Japanese ecommerce giant Rakuten invested, are good examples of this trend. Based in Hamburg,Kreditech targets the millions of unbanked people, studying their digital footprint to offer them credit, where applicable, either directly or through third parties.
In addition to users’ accounts on Amazon, eBay and Facebook, Kreditech also analyzes certain digital behaviors that give clues regarding solvency levels. For example, the amount of time spent checking the credit conditions, the use or lack of use of capital letters and the place where users go online matching the place where they say they live or work. And these technologies are not just useful for analyzing the risk of a credit, but also for detecting possible fraud and online impersonations.
Most data from social networks are collected directly or through their APIs. But what about all the people who don’t have accounts on social networks? “The fact that someone doesn’t have a digital footprint is important data,” Luna de María explains. “They don’t want to share information and that in itself is information. In those cases, we use classic solvency analysis models, but it’s important to keep in mind that just a email address provides a lot of information.”
Privacy is more complicated in the 21st Century, but like all social changes brought about by technology, the digital footprint has both positive and negative aspects. Many of those who were not previously part of the banking system can now access credit thanks to their presence in social networks.