top of page

AI Case Study

Tala extends loans to 150,000 Kenyans based on their mobile phone data with a 90% repayment rate on $50 micro-finance loans using machine learning

Tala operates in developing economies where credit is not easily available for majority. By using an app to track communication habits, phone usage, social media channels etc they analyse the likelihood of customer repaying the loan. In Kenya they start with a loan of $50 which has a repayment rate >90%. Over time based on payment history the credit worthiness is adjusted.


Financial Services


Project Overview

"Tala introduced a consumer lending app that underwrites customers in real time using thousands of alternative data signals. Anyone with an Android smartphone in the markets can apply for a loan and receive an instant decision, regardless of their financial history. Tala provides fast, personalized loans to approved borrowers and help customers build a digital credit history, or financial identity, over time.

After downloading Mkopo Rahisi and allowing the app to access their device, they receive a decision on credit nearly instantaneously, based solely on data from their daily life and habits." 

From microfinancegateway, "Tala’s customer engagement leverages customer data to provide a personalized financial
experience via a sophisticated mobile application and through social media channels like Facebook. Through the app, customers can manage all aspects of their account, including checking balances, making payments, or accessing support through an in-app messenger which promises a response within 24 hours. They can also track their customized ‘Tala credit score’, set financial goals, and use personal financial management tools. "

Reported Results

"On loans that average $50 with an interest rate of 11%, Tala has logged a repayment rate of better than 90%. To date it has loaned almost $20 million to more than 150,000 people. On revenue of $1.5 million in the last year, it had a profit of more than $500,000. In this interview, which has been edited and condensed, Siroya describes how she pioneered this new form of micro-lending."




Sales Operations


"Only 31% of the adult population globally is covered by credit bureaus. In markets where bureaus are limited or nonexistent, Tala can use nontraditional data to can score up to 2X that population – introducing 3 billion consumers to the global market."



"10,000 data points from the phone such as data from a user’s social networks, mobile money ledgers, and GSM data, calling habits, sms habits, call time, social media channels like Facebook"

For eg: data shows us that good borrowers tend to have strong relationships with a wide network of people which can be measured from communication habits

bottom of page