Success Stories

How Creditt built an AI to improve loan defaulting rates

Creditt is a micro-lending company that uses Obviously AI to predict loan repayment in minutes, reducing their loan processing time by 5X. Read more to learn how they do it.


AI model predicts loan repayment for 20K applications per mo.

Time Saved

Reduced loan decision time from 20 minutes to 5 minutes

Reduced Backlog

Saves 208 calendar days of backlog each month


Creditt is a micro-lending and financing company that addresses the credit needs for underserved communities through their automated mobile app.

Data Sources Used:
Data Merging
Data Enrichment
Statistical Work

The Problem

Creditt receives a massive flow of 20,000 loan applications per month. Each of these applications take 20 mins to process manually. This means every month, they add a backlog of 18,000 loans that keep piling up for the month after. Creditt identified that using AI to predict loan defaults would massively reduce their backlog. However, they didn’t have a massive team of data scientists in-house who can build, deploy and automate these AI models. Additionally, their attempts to work with services companies and other AI modeling tools resulted in more complicated and less accurate results.


Creditt turned to Obviously AI to build out their AI models. They simply signed up to get started and were building models within minutes. They also worked with a dedicated data scientist who provided industry best practices and suggestions to further improve their model accuracy. After which Creditt’s team incorporated these models into their apps so each time a loan application was submitted by a user their default prediction was automatically made by the model and served to the underwriter. The team was able to go from data to ROI in weeks without any AI expertise.


Using Obviously AI, Creditt automated the loan application evaluation process. This reduced time to process each loan application from 20 minutes to 5 minutes. In turn, saving over 208 calendar days of backlog per month.


Obviously Al is a no-brainer tool for predictive analytics. Easy to integrate and predict outcomes without any knowledge of machine learning.

Nama Parikh
Creditt India

Get Started Now

See how no code machine learning can transform your business and change how you make decisions.