AI predictive analytics, machine learning and the use of alternative data are key to making improvements in many areas of lending to small and midsize businesses, but there are many considerations to ensure SMB lenders are positioned on the path to progress.
A key challenge for SMB lenders is understanding which alternative data elements provide the right signals relative to what they are trying to achieve as a business. Lenders can procure reams of third-party data available in the marketplace, but this process must be conducted with a thoughtful understanding of what’s going to differentiate their organizations from the competition. Otherwise, they can be stuck in a data lake with no sail for years and not get the answer they were hoping to find.
It’s vital that SMB lenders initiate business accountability by answering the following questions:
- What are we trying to answer relative to predictability and/or information around credit decisioning that we can’t answer today?
- How does alternative data augment that, and what are the right alternative data elements that give us the appropriate signals relative to what we’re trying to achieve as a business?
- How do we want to differentiate ourselves in the marketplace, and what alternative data will best supplement our first-party data and allow us to achieve that differentiation?
With artificial intelligence and machine learning, it is important to take a holistic approach across the entire funnel. AI can help lenders understand the predictability at each inflection point in that funnel, from an early sales lead to application approval based on submission characteristics.
There are always tradeoffs in building your own solutions or leveraging packaged software, or software-as-a-service solutions. Two key considerations are speed-to-market and the lender’s points of differentiation relative to its competitors.
In today’s AI and machine learning world, if a lender does not have technology platforms and systems that allow it to move at pace in line with market expectations, it will fall behind with systems that are inhibiting to the business.
This is why it’s important to take a disciplined approach to the need to have things that they can configure outside of system software changes versus where they want system software changes. For example, if they have to make changes relative to pricing and credit risk scoring, they can make those in a rapid timeframe as market conditions change, rather than initiate a software development build. If a lender’s software solutions don’t facilitate these changes, then it will quickly become a laggard.
Bottom line: An SMB lender’s technology choices become an integral part of its cost-to-serve model.
Lenders need to understand how their systems facilitate how they actually serve customers by providing the products and services that they’re looking to use. The prospect of getting to market faster is not necessarily the end game if it costs three times what it should and degrades margins and profitability. For this reason, banks and credit unions should carefully weigh building custom applications versus leveraging SaaS solutions and technology.
By understanding technology’s impact on the cost-to-serve model, SMB lenders need to ask themselves the following questions:
- How do we continue to drive margin growth?
- How do we continue to drive efficiency but still have the human touch only where it adds value?
- How do we leverage machine learning to drive better predictability, ensuring that we are both going after the right opportunities for merchants, and that we’re facilitating the right credit and decisioning processes?
Leveraging those answers all in concert and thinking about technology systems from a profitability standpoint is very important, because technology can be a fairly large cost component of any organization if not properly applied. In contrast, strategic thinking in this area ensures that business considerations become an integral part of leveraging technology effectively.
Done right, SMB lenders can transition from traditional policy-based approaches to those that leverage predictive, explainable and scalable machine learning to radically improve the speed and accuracy of loan approvals while ensuring margin growth.
John Pesavento is vice president of technology for Reliant Funding.
