TrustingSocial, a fintech startup headquartered in New York, is looking to tap into a trillion dollar market opportunity by determining your creditworthiness with social and mobile data.
By using home-designed Big Data and deep learning technologies, TrustingSocial aims to find out how good of a borrower you are by using data collected from popular social networks and telecom operators.
The startup has partnered with Vietnam’s largest mobile network operator Viettel, and ran a pilot only to find out that its prediction model was „50% more accurate than the best model in the world,“ according to co-founder and CEO Nguyen Nguyen, a Vietnamese native and former credit risk researcher at Barclays.
The companys deep learning technology enables the company to detect consumer profiles and behavior patterns from a trillion of records provided by telecom operators. Furthermore, its credit graph computing technology allows it to detect social linkage.
Designed to „make lending faster, cheaper and friendlier,“ TrustingSocial’s credit scoring technology seeks to reach three billion of „invisible customers“ from emerging markets.
This group of consumers is often considered unscoreable by the traditional credit system because of a lack of credit history. „Banks just ignored them,“ Nguyen argues.
Pitching his startup at SWIFT’s 2015 Startup Challenge, he explained:
For example in Vietnam, only 5% of the population has credit history. All the banks compete; offer all kinds of products to that 5% and they left out 95%.
„Our solution is to provide 3 billion credit scores based on mobile and social data.
In developing countries, such as India, Mexico, the Philippines or Colombia, social media has become an alternate method of deciding whether to give a loan to somebody or not, according to Pinar Yildirim, a marketing professor at Wharton.
„The reasons why they find these countries suitable is, first of all, there is a lack of financial access, and yet a huge number of people are moving into the middle class,“ Yildirim said on the Knowledge@Wharton show on Wharton Business Radio.
Lending companies „have these somewhat irrelevant, almost unacceptable criteria for you to obtain a small loan. […] Without looking at anything else, you just immediately disqualified someone from obtaining access to credit,“ she argued.
Because these populations have no credit history or no history of financial payments whatsoever, people need to look for alternate ways of finding data about one’s character; and considering the high mobile phone penetration in developing countries, social media makes perfect sense.
„In an environment of this sort, what you are going to end up with are very detailed data about people’s mobile phone use,“ Yildirim said. „And because you know mobile phone use, you end up having information about social network [use] as well.“
For a smart company, this creates an alternate way of getting data about people’s personalities.
But when it comes to collecting users‘ social and mobile data, the concern of one’s data protection arises. Well, to deal with information privacy, TrustingSocial said it has invented a „zero-knowledge data sharing protocol“ that makes „all the information transferred between banks and telecom operators and us, [anonymous].“
„Telecom operators transfer usage data to us but we don’t know who the practical consumer is, and the same goes with bank data,“ Nguyen said.
Qualified as one of „the most promising fintech companies in Asia,“ TrustingSocial will compete in SWIFT’s Innotribe Startup Challenge Grand Finale taking place at Sibos on October 14, in Singapore.
Launched in 2011, SWIFT’s Innotribe Startup Challenge aims to bring together global innovators and investors, strategist and influential decision-makers from leading financial institutions across the global. The competition seeks to select the most impressive fintech startups and help them benefit from a unique networking opportunity with banks and VCs during regional showcases.
Watch TrustingSocial’s pitch at the 2015 Startup Challenge at Next Banking Singapore in May:
For more Information about Fintech Singapore and South East Asia please subscribe to the Fintech Singapore Newsletter
[…] company claims that its prediction model is “50% more accurate than the best model in the […]