Speaker "Divya Beeram" Details Back
-
Name
Divya Beeram
-
Company
Intuit
-
Designation
Data Scientist
Topic
Predicting small business financial needs and finding the right lenders
Abstract
At one point or another, every small business needs financial funding to grow, or even just to keep their business running. For small businesses, it’s a struggle to get short-term funding. Traditional banks often lend based upon the credit worthiness of the business owner combined with financial factors of the business itself. The process of pulling together the key documents can be tedious. The loan application process on average can take a business owner 33 hours of time to complete. Then there’s the question of when small businesses actually need funding? Too often small business owners wait too long to begin the long and complicated loan application process.
At Intuit, we have partnered with lenders to help small businesses get the funding they need, when they need it, to help their businesses thrive. We help the small businesses unlock their data in QuickBooks (small business accounting software) and use it to quickly get through the application process. By leveraging their accounting data combined with a deep understanding of how specific financial lenders make decisions, we are able to help small businesses find the right lenders and quickly get the funding they need.
To accomplish this, we must accurately identify which businesses are looking for/need funding at any given point of time and find the right lender for the right business.
This talk will focus on how we have leveraged the transactional, firmographic data in Quick Books to (a) build machine learning models to predict the need for funding (b) implement the model in production and (c) build a feedback loop for model improvements. We will also touch on common pitfalls and lessons learned throughout the process.