Download PDF
DataRobot > Case Studies > Kiva Uses DataRobot to Increase Microloan Funding Rate
DataRobot Logo

Kiva Uses DataRobot to Increase Microloan Funding Rate

Technology Category
  • Analytics & Modeling - Machine Learning
Applicable Industries
  • Finance & Insurance
Applicable Functions
  • Business Operation
Services
  • Data Science Services
The Challenge
Kiva is a financial services nonprofit that uses crowdfunding to underwrite loans for people who are underserved by traditional channels. The World Bank estimates that approximately 1.7 billion people are unbanked, meaning they do not have access to financial services offered by retail banks. This leaves many people without access to the financial instruments that much of the world takes for granted, such as credit cards and loans. Alternative banking methods tend to have high fees that can put them out of reach for the people that need them. This lack of capital hinders economic growth, opportunity, and equality in the places that need it the most. The key to Kiva’s mission is to ensure that those who apply for loans are successfully funded.
About The Customer
Kiva is an international nonprofit founded in 2005, with a mission to “expand financial access to help underserved communities thrive.” Kiva accomplishes its mission by offering crowdfunded loans through its international network. The loans are used in various ways such as helping women start and grow their businesses, enabling farmers to purchase equipment, and providing students with loans to further their education. The organization is based in San Francisco, CA and operates in the nonprofit and financial services industry.
The Solution
With the goal of ensuring successful funding for loan applicants, Kiva partnered with the DataRobot AI for Good program to develop a system to promote loans that aren’t being viewed and are at risk of going unfunded. Using data about each loan, including daily popularity, the Kiva team used DataRobot to build more than two dozen models that predict which loans will be funded each day. This solution is built to integrate directly with the Kiva website, reordering the loans daily in a way that provides visibility to any that are at risk. The solution resulted in the promotion of loans in geographically diverse regions including parts of the world that are sometimes overlooked by funders.
Operational Impact
  • The solution integrated directly with the Kiva website, reordering the loans daily to provide visibility to any that are at risk.
  • The solution promoted loans in geographically diverse regions including parts of the world that are sometimes overlooked by funders.
  • This solution is the first large-scale machine learning model implemented by Kiva for this problem.

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.