In a Nutshell: Accessing credit in developing economies can be a long, difficult task that can affect a borrower’s well-being and financial success. LenddoEFL combines big data analysis and psychometric evaluation to create alternative credit risk assessments that help people in emerging markets access vital financial services. The company develops a comprehensive understanding of a consumer’s creditworthiness by examining behavior as simple as how often they charge their phone’s battery or how they play an interactive game. By leveraging these tools and keeping up with emerging sources of data, LenddoEFL is striding toward its goal of providing 1 billion underserved individuals with access to credit that would otherwise be out of reach.
Credit is hugely important to people around the globe. You need it to obtain housing and higher education. You need it to start a business. You need it in case of emergencies and other unexpected expenses.
But in emerging economies, credit may not be accessible to many people. According to the World Bank’s 2017 Global Findex, 31% of the world’s population doesn’t have an account with a financial institution or a mobile money provider.
“We still have 1.7 billion people on the planet who don’t even have a basic bank account,” said Amie Vaccaro, Director of Marketing at LenddoEFL. “Only 11% of people around the world borrowed from a formal financial institution in the last year.”
LenddoEFL’s mission is to change the game for the 89% who don’t even have access to a credit card. By leveraging machine learning and psychometrics to provide an alternative way of assessing risk, LenddoEFL is making credit accessible to people who otherwise wouldn’t be able to obtain any.
“It’s really hard to live a full life without access to credit,” Vaccaro said. “Essentially, the mission of the company is to provide 1 billion people with access to life-changing financial services.”
Evaluating 12,000 Variables in Three Minutes
LenddoEFL developed from the merger of Lenddo and Entrepreneurial Finance Lab (EFL). Lenddo specialized in leveraging machine learning and AI for analyzing digital footprints and data. Over the course of four years, the company collected, analyzed, and processed data to develop a nontraditional method of computing credit scores.
EFL grew out of a Harvard Center for International Development initiative aimed at applying psychometric assessment as low-cost credit screening tools. The Lab’s purpose was to facilitate entrepreneurship in emerging markets by closing the information gap on borrowers whose creditworthiness could not be assessed by traditional means.
Today, LenddoEFL combines data analysis and psychometric profiling to offer financial inclusion to underserved markets around the world.
“The big difference between us and a traditional credit score is that we’re designed for these low-information scenarios,” said Vaccaro. LenddoEFL’s analyses complement traditional credit bureau data, and combining the two creates an even more comprehensive, predictive image of an individual.
The LenddoEFL system — which evaluates 12,000 variables in three minutes — has assessed 7 million potential borrowers through more than 50 client institutions in 20 countries. The screening process has resulted in over $2 billion in capital loaned, 50% greater approval ratings, and a decrease in defaults.
“Together, we’re providing banks around the world access to these new types of data to help them make better decisions,” Vaccaro said, “and ultimately approve more people for loans and for credit products.”
Risk Assessment Through Digital Interactions
There are 1.5 billion unbanked adults in the world, but more than 66% of them have mobile phones. One way LenddoEFL helps these individuals get access to credit is by analyzing their digital interactions and data.
“We take these various behavioral and digital footprint data sources and use them to quantify risk,” Vaccaro said. “There’s a lot of information about who you are, how you behave, and your personality just in how you interact with your phone.”
One example is battery usage. “Generally, folks who keep their phone charged tend to have lower risk,” Vaccaro said. Another aspect is the completeness of contact information; when users enter both first and last names of contacts, it’s a good sign of lower credit risk. LenddoEFL can also use online profiles, connections, and interactions to assess risk and verify identity.
For both local data and information on the cloud, data privacy and security is LeddoEFL’s top priority. “Every piece of data we touch about a person is consent-based,” Vaccaro said. “People can opt out at any time as well.” LenddoEFL accesses applicants’ personal information only once during the assessment and never shares it with lenders or third parties.
Personality as an Indicator of Creditworthiness
You may be wondering about the remaining 33% — the unbanked and the unphoned. LenddoEFL has a solution for them, too.
“Somebody who doesn’t have any sort of digital presence could take a 20-minute assessment on a tablet, smartphone or feature phone” Vaccaro said. “This is where psychometrics really come in.”
“Even if someone is illiterate, the loan officer can read the questions to them and get their answers,” said Vaccaro. “It’s really designed with the full range of access in mind.”
The assessment is a series of interactive modules that applicants play like games. “It was meant to be somewhat fun and also gives us ways to learn about somebody,” Vaccaro said.
One benefit of this format is the built-in possibility for “cheating,” or gaming the system. “We know people are going to try to get their best score, and we want to understand how they’re gaming it,” said Vaccaro.
For LenddoEFL, the metadata describing the user’s interactions is just as important as the explicit answers they provide. In particular, information on how long it takes a user to respond, whether they redo an exercise when given the opportunity, and how seriously they’re treating each exercise all help to create an accurate profile of the individual.
“Essentially, each of these data sources helps us understand personality,” Vaccaro said. Through the interactive assessment, LenddoEFL can gather information about an applicant’s confidence, conscientiousness, honesty, risk tolerance, and gratification.
“We can learn about somebody’s personality and their behavior,” she said. “That can help us predict risk and help sort between the applicants when there’s no other information available.”
Changing the Face of Personal Finance
In the days of the Commodore 64, it may have been difficult to believe something called the internet would one day revolutionize our lives. And around the turn of the millennium, understanding how social media platforms, like Facebook, Twitter, and Reddit would eventually change the face of the internet — or that we’d be able to use them on our phones — was well beyond most people’s comprehension.
Now, LenddoEFL is using these advances in technology to change the nature of finance. But the company thinks its work is only beginning.
“We’re always looking for those next sources of data,” Vaccaro said. Whatever form those sources take, they’ll bring LenddoEFL closer to its goal of helping 1 billion global consumers gain access to much-needed financial services.
“New sources of data that people are creating can help us learn about them,” Vaccaro said, “and give them more opportunities for credit.”