Grade-based loans? See me after class

New idea to determine credit score

Those who believe SAT scores don't matter after college acceptance letters come in may want to think again.

A company called Upstart is using a variety of unconventional factors to determine if consumers are eligible for lending. The San Francisco start-up is targeting millennials between the ages of 22 and 34 who may have had a hard time qualifying for loans.

Upstart was started by ex-Google employees with the belief that consumers with certain characteristics will be more responsible if they see debt as an obligation. It analyzes factors such grade point average, SAT score, college attended and even college majors.

"These variables, they are extremely predictive on whether someone is likely to pay a loan," Paul Gu, co-founder and head of product said in an interview with CNBC's "Power Lunch."

"We think every person we're helping with our loans is another person we're getting out of credit card debt."

Jamie Grill | Getty Images

In the past year, Upstart has issued nearly 10,000 loans worth $143 million. Most recently, the company raised $35 million in Series C round funding led by Third Point Ventures.

Read More Student debt pushes off financial milestones

Millennials have lower credit scores than older generations, according to a study by credit bureau Experian, which is largely due to them having a more limited credit history than older generations. According to Bankrate, 63 percent of millennials don't even have a credit card, compared to just 35 percent of adults over 30.

The unorthodox credit model, however, is drawing concerns from some who see a potential for lending discrimination.

"This whole concept is ridiculously subjective, even though proponents try to claim it is objective," Lynnette Khalfani-Cox, personal finance expert and author of "Zero Debt," told "Power Lunch."

"There is a massive amount of assumptions built into these scoring models."

Read MoreSmart strategies to help millennials pay off debt

Despite Upstart using algorithms to determine who is eligible, Khalfani-Cox still sees the model fraught with pitfalls.

"What about all the doctors and lawyers with tons of student loan debt who don't repay? They had great test scores and grades but they often borrow six-figure sums to pay for med school and law school."