Tala is a data science company that provides micro-loans ranging from $10 to $500 to low-income individuals in emerging markets. Only 69% of adults worldwide (63% in emerging economies) have an account at a financial institution or through a mobile money provider , 1.7 billion adults remain unbanked . Tala enables people to access personalized, uncollateralized loans through a smartphone app, regardless of credit history.
Profit through an innovative use of data
In 2016 Tala had provided $20 million in loans to more than 150,000 people . In 2019, it made $1.5 million in revenues, for a profit of $500,000 . The startup has raised $220 million from private investors and has a valuation of $1 billion . “The idea of lending purely based on the data available on a consumer’s mobile device was completely unproven — no one had ever done it” said one investor . Many financial service providers operating in developing countries have data, but they lack the capacity to store and analyze this data.
How Tala uses data to build credit scoring and a financial identity
Tala’s scoring model relies on more that 250 data points divided in 2 categories: Android device data, and behavioral data. This data are used for credit assessment and includes name, gender, age, purpose of the loans, proof of utility or phone bill payment, information on the borrower’s online behavior, etc. Behavioral data includes “how customers interacted with the app, if they read the terms and conditions, any mistakes while typing basic biographical information” . The data even reveal where the applicant goes during the day. Device data such as type of mobile phone used and year of the operating system allow the identification of the borrowing capacity while mobile bill payment history and behavioral data help assessing the likelihood of repayment. This is because, according to Tala’s approach, “a person’s routine habits are more meaningful than traditional credit scoring” .
Speed and accuracy of the Tata’s model
The disbursement process is fast, and it supports borrowers in need of emergency loans. After downloading the app, the borrower responds to 8 questions and is immediately scored. After the scoring he receives a loan offer and, if he accepts it, he gets the money in a mobile wallet. 85% of customers receive cash in their wallet in 2 minutes . The scoring model designed by Tala is so accurate the repayment rate of loans is higher than 90% .
Beyond credit scoring
The use of data analytics is not only aimed at building a credit score. Through data collection Tala is building a “financial identity”  for people without any form of identification (such as identity documents or voter cards), who are normally excluded by traditional financial services.
Challenges and opportunities
A primary factor of concern is that, although Tala reports that data are encrypted and personally identifiable information are not shared with third parties, issues on the data protection might arise . Next, Tala faces a high competition in the financial services space. Currently, the competition is increasing due to the high popularity of digital financial services, which is partially caused by the COVID-19 pandemic which is forcing digitalization of payments. Finally, another challenge, that could also represent an opportunity, is that the present loan amount offered by Tala.
 Demirgüç-Kunt, Asli; Klapper, Leora; Singer, Dorothe; Ansar, Saniya; Hess, Jake. 2017. The Global Findex Database 2017. Measuring Financial Inclusion and the Fintech Revolution (World Bank Group). Available at: https://globalfindex.worldbank.org
 Shetty, Sameepa. 2020. “Start-up uses mobile data as a credit score for the global unbanked”in CNBC, Jan 3, 2020. Available at: https://www.cnbc.com/2020/01/03/start-up-uses-mobile-data-as-a-credit-score-for-the-global-unbanked.html
 Adams, Susan. 2016. “How Tala Mobile Is Using Phone Data to Revolutionize Microfinance”in Forbes, Aug 29, 2016. Available at: https://www.forbes.com/sites/forbestreptalks/2016/08/29/how-tala-mobile-is-using-phone-data-to-revolutionize-microfinance/?sh=304d2dee2a9f
 Shetty, 2020, cit.
 Adams, 2016, cit.
 Shetty, 2020, cit.