Daniel Hinkson and Ryan Thorpe are the data science department at Crest Financial, an alternative lending firm providing borrowers with “No Credit Needed” financing.
One of the nation’s largest and fastest growing “No Credit Needed” lease-to-own companies, Crest Financial offers technology-driven instant approvals for items like furniture, appliances, tires, and personal medical devices on transactions up to $5,000. Crest is committed to providing more opportunity for customers, more success for retailers, more service to the community, and more growth for employees than any other company in their industry.
Underwriting decisions are based on basic data supplied by the customer, and supplemented with payment histories like rent and utilities. Crest relies on predictive models created by Daniel and Ryan to help present a solid option to consumers with no access to traditional credit or financing.
Despite their heroic efforts, Daniel and Ryan were hampered by traditional data science methods, where building and testing machine learning models manually often took months, and deployment delays were common. To keep up with the ever-increasing workload, and to deliver faster, more accurate lending decisions, Crest needed a vastly-improved predictive modeling solution that would be not only accurate, but lightning-quick as well.
Looking for help, Daniel and Ryan sent out the DataRobot Signal.
Daniel and Ryan worked with the DataRobot team to evaluate the DataRobot platform on Amazon Web Services (AWS). Crest wanted to see how DataRobot on AWS could help them improve risk analysis, better identify marketing targets, and improve and expedite detection of fraudulent applications.
The results from less than an hour of proof-of-concept testing showed DataRobot models delivering better accuracy than the hand-built models developed over many weeks. Crest Financial was all in, quickly becoming a part of DataRobot's data science superhero universe.
With DataRobot, Daniel and Ryan quickly discovered their data science superpowers and transformed into data science superheroes. With their newfound abilities, Daniel and Ryan have the power to develop and deploy predictive models with lightning speed. Now they can:
- Pinpoint the right type of consumer for Crest in a high-risk and highly-competitive market.
- Quickly identify instances where applicants are attempting to obtain financing through fraudulent means.
- Accurately predict the likelihood of default for applicants.
|Traits & Capabilities|
|Traits & Capabilities||Data Scientist Mortal||Data Scientist Superhero|
|Bottom Line Business Impact|
|Bottom Line Business Impact||$100K+ per year||$M's+ / Year|
|Model Engine Tested|
|Model Engine Tested||No||200,000,000+ times|
|Standardization||When there's time||Every time|
|Multitasking||Single model||Multiple models|
|Package installation and dependency management|
|Package installation and dependency management||Manual||Not needed|
|Risk of coding errors|
|Risk of coding errors||Substantial||None|
|Popularity with business users|
|Popularity with business users||Somewhat||Very|
|Business user engagement|
|Business user engagement||Too busy||Always available|
|Helped with guardrails|
|Helped with guardrails||None||Yes|
|Skills Needed||Significant - PhD-level||PhD not required|
|Model deployment effort|
|Model deployment effort||Significant||Effortless|
|Model iteration||Manual & slow||Automated|
|Helped by other Data Scientists|
|Helped by other Data Scientists||No||85+|
|Algorithms and libraries|
|Algorithms and libraries||A comfortable few||Many|
|Number of models / month|
|Number of models / month||Very Few||Thousands|
|Model Transparency||None||X-Ray Vision|
|Explainable Models||'Black Box'||Automatic|
Next steps for Crest Financial?
DataRobot will allow Daniel, Ryan, and the Crest management team to continually refine models and enhance customer experience and retention. Since DataRobot provides an easy-to-use platform to iterate and update predictive models, Crest Financial can stay ahead of the competition by rolling out new models faster than ever. And with Crest’s help, credit-challenged consumers will find themselves able to get what they need and when they need it the most.
“What DataRobot was able to accomplish in the first hour was more thorough and accurate than models we had built over the prior month.”