Nova Credit enables SoFi to verify customer income through bank transaction data and expand financial access to the informal economy

Nova Credit and SoFi partnered to develop an income identification model using bank transaction data, enabling SoFi to reduce its underwriting costs through automation, and eventually, reduce loss rates given a more accurate and holistic income assessment of the consumer. SoFi is now also able to serve a segment of consumers who do not have traditional employment, such as gig workers, self-employed individuals, and retirees.

Previously, income verification would only capture a consumer’s primary income source, thereby not providing a holistic view of a user’s credit-worthiness, also resulting in costly, burdensome, manual review processes for many ‘non-traditional’ applicants. 

SoFi’s Chief Credit Officer, Ratinder Bedi, said ”In theory, bank transaction data gives you tremendous visibility into a consumer's ability to pay, but in practicality that data is extremely complicated and not easy to make sense of.”

For SoFi consumers where automatic back-end income verification fails, Nova Credit now provides those users an opportunity to verify their income by linking their bank account. In near real-time, the model is able to determine whether an applicant’s gross income is in line with their stated income, and the loan application can proceed to the next step of the approval process. A typical applicant may have 18 deposit streams, but just 2-3 income streams. The power of the model is in its ability to identify and distinguish these income streams. Consumers would otherwise be asked to upload paystubs, which lenders like SoFi would typically have to verify through time-intensive and costly manual processes.

Mr. Bedi, SoFi’s Chief Credit Officer added “Nova Credit's income verification tool has delivered excellent, tangible results for our business – higher loan conversion, a shorter time to fund, and ultimately a great member experience because the process is so seamless”

Read the case study