The Case for M-Score at MOMBO Sacco
MOMBO Sacco M-Score is a scoring engine that applies unconventional rich data to the core underwriting model thereby expanding the lending space for individuals that were previously considered “thin file” or “risky” when Binary credit data was used to evaluate them by established financial services players.
Below are the key pillars of the MOMBO Sacco M-Score model:
- Transaction behavior – this consists of Utility payments, spend at Till/ Paybill, payment behavior mainly from mobile money history
- Demographic and Bureau data – a combination of demographic information with product holding and repayment behavior for existing borrowers.
- Geo-location data – GPS information coupled with financial transaction behavior.
- Personal finance Management – MOMBO Sacco savings history, repayment history, transaction history.
- App-based data – MOMBO app usage, interactions, texts, e-mails, retail receipts, collection texts to assess the creditworthiness of the customer.
- Machine Learning and automation of backend through robotics – Continuous feedback for model efficacy and model finetuning.
MOMBO Sacco for instance uses mobile phone data and e-commerce sales as additional data points for analyzing consumer behavior. Parameters such as long call duration, till spend, frequent high-value mobile top-ups, transactions to certain paybills/ are generally considered as positive indicators, while low-value top-ups, missed savings etc are associated with lower credit scores. We have recently been experimenting with light question/ answer approaches that gauge a customer’s intention to pay—a technique that is especially valuable in the case of thin-file/no-file customers, where other data is scarce, and the customer has saved for a few months.
The use of large data sets that are analysed computationally to reveal patterns, trends, and associations, especially relating to human behavior and interactions has changed the game for us in 4 key ways:
- Introducing transparency to the transactions. The data obtained from customers helps us customize the price and offering for each individual while on the savings side, we can run discounts by having the system pick early savings for instance, or early loan repayments which earn customers 50% discount on the interest rate charged.
- Eliminated old multiplier approach to what the customer can afford has removed the incessant liquidity challenges faced by traditional Saccos. The customer only qualifies for what they can afford.
- Behavioral patterns picked from the large data sets have enabled us to pick patterns over the years and successfully deploy an offer management system. For instance, in the period towards December, customers have exciting offers to help them save more for holidays or schools fees and at the same time qualify for better loan sizes.
- Consumer behavior has enabled us to customize bite-sized daily saving goals for our customers thereby instilling discipline in their savings culture. Data analytics has assisted in driving customer advocacy and thereby competitive advantage for MOMBO Sacco.
By Mombo Sacco | March 30, 2022 | 4 min read |
0