Automated underwriting system for SMB lenders
Traditionally, the loan underwriting process has been a slow, cumbersome affair. But now, lenders are increasingly adopting automated underwriting systems to streamline the process and expedite loan decisions.
By
Traditionally, the loan underwriting process has been slow and clunky with multiple data sources and lots of paperwork to sort through. Lately, more and more lenders have adopted automated underwriting systems, including Freddie Mac, which announced automated underwriting capabilities for mortgage lenders in 2022. While mortgage lenders still haven’t adopted a fully digital end-to-end solution, they have benefited from decreased costs and faster processing times.
Automated underwriting can be equally beneficial for SMB lenders.
How automated underwriting works
At the most basic level, underwriting automation entails using technology to collect and analyze information about a borrower to make a loan decision.
An automated underwriting system will verify firmographic and demographic information, along with third-party data, including identity verification, tax documents, know-your-business compliance checks, credit bureau checks, and account verification.
Once the data has been collected and analyzed, the system can finalize approval of the loan within 15 seconds.
Financial institutions may choose to implement automated underwriting in a number of ways including:
- Using automation to source data, but still allow a human to review and give final approval.
- Using automation to source data and recommend a decision.
- Relying solely on automation to make a decision.
Benefits of automated underwriting
The benefits of automation in banking are vast, including:
- Faster processing times – Approving a loan in seconds, rather than days, provides a better experience for the borrower and frees up employee time to focus on activities that can have the biggest impact on the financial institution.
- Better outcome predictions – Multiple studies across auto, mortgage, and commercial lending have found that machines can more effectively analyze multiple datasets to predict outcomes than humans can, lowering the risk of default and decreasing the cost of bad debt.
- Less bias – Automation has been credited with reducing human bias in the loan approval process, but modern underwriting systems provide additional benefits to traditionally underserved communities by pulling in data sources that better reflect the health of the business, rather than relying on the business owner’s personal credit score.
- Expanded addressable market – Due to time and cost constraints, banks typically focus on servicing larger loans. By automating processes, banks can reduce the time and money spent prequalifying and underwriting a loan, making it possible to service the SMB market—which makes up a large part of the bank’s depositors—and access millions of new revenue opportunities.
- Wider distribution – Embedded finance and banking-as-a-service channels are creating new frontiers in lending. Banking-as-a-service is projected to reach a market size of $11.34 billion by 2030. However, along with those channels comes new requirements for the re-use of existing borrower data, faster time to offer, and faster time to fund. Automation is becoming a necessity, particularly in these times.
Documenting policies and sticking to them
Banks pride themselves on having well-documented policies. In fact, many banks have taken important strides to sharply reduce—and even eliminate—the elements of subjectivity in decision-making. However, in practice, lenders often have exceptions. But what if, rather than designing a process to accommodate the exceptions, we designed a process to optimize for the vast majority of applications that can be accurately measured and adhere to a written policy?
Machine vs. human decisions
Of course, like the adoption of any new technology, automated underwriting has produced a debate around when/if a machine truly is a better option than a human. A human, for example, can better assess nuance than a machine. Does this then give the human the advantage in better predicting outcomes? In actuality, studies have found that machines are better able to assess risk for borrowers with lower credit scores or a prior history of bankruptcy.
Users adopting the technology also need to trust that the decision engine is accurately assessing risk profiles. To help alleviate those concerns, underwriting systems can be adjusted to match the bank’s risk profile.
Going beyond underwriting
Automation in banking has a much bigger role beyond underwriting. It can be used to pre-qualify borrowers and market to them, creating a much more attractive offer for the borrower and better ROI for the lender’s marketing and loan officer teams.
In this scenario, the loan origination system pulls in transaction data from the bank’s system and third-party sources, classifies those transactions, tabulates results to develop a profile, and then runs that profile against a predetermined transaction policy. Based on the results, the bank can then send a pre-qualification offer to the customer.
Bottom line
There is a huge opportunity for the lending industry to become more efficient, reduce risk, and capitalize on the underserved small business market via automated underwriting. Banks that adopt this technology early will have a key competitive advantage over other institutions as automation in banking becomes more mainstream.
Are you interested in lending automation? Learn more in this ebook, Automated Lending: A Mandatory Upgrade.
