5 problems lenders have adopting automation and how to fix them

Surprisingly, many lenders with a long history of AI automation have yet to find their stride. Large lenders found early success with declarative decisions (rules-based decisions) for many mission-critical processes, such as determining pricing and underwriting, but struggle with machine learning. Meanwhile, mid-sized lenders face obstacles implementing AI solutions because their packaged products can’t be customized, and layered solutions typically don’t fit naturally into their everyday workflow. 

Regardless of the size of your organization, the principles below have helped many lenders overcome roadblocks to AI automation success. 

Working with many new hires
Loan officers are busy and many are new to the industry, having been recruited from retail sales positions. Loan officers need fewer distractions and a safety net while working with customers. It’s tempting to add yet another screen or an external piece of software, but that’s not always the quickest path to success. 

Take time to follow the journey and look for solutions that address the most urgent problem. Timing, accuracy, and fit can all be improved but only if they do not cause harm to the natural rhythms of a journey. Seek out assisting technology that will make loan officers’ jobs easier, while simultaneously pleasing customers with faster-than-expected turnaround times. Digital transformation is about everyone’s experience — not just that of the customer.

Automation must be seen as a strategy to work seamlessly with people, rather than as something that must be worked around or — worse — as a penalty. In other words, automation should be seen as part of natural workflow in a core business process. Be careful about tools and experiences that require yet more complexity and instruction. Cognitive overload is the enemy.

Learning curves and data overload slow path to modernization 
While automation in lending dually serves customers and employees, unfortunately change isn’t always welcome. For example, implementing modern technologies and new processes can cause confusion and frustration among employees who already feel stretched. Yes, it’s challenging to fix the ship while under sail. It adds risk, but to win, the organization must take take steps forward. 

Starting small isn’t always going to help. Targeting a strategic problem that matters will keep the teams oriented to a north star. While right-sizing remains important for the first project, it’s not the only concern. The impact must be meaningful.

Lenders don’t always trust AI 
Loan officers might be wary of tech that, on the surface, can do their job. However, AI automation doesn’t eliminate the need for loan officers. Rather, it enhances their decision-making capacity, making them more productive and less prone to error. In fact, greater transparency about what the automation is doing dispels anxiety and delivers on other fronts, such as fairness and compliance.

Lenders should prioritize solutions rooted in explainable AI. This capability is becoming increasingly critical for regulatory compliance in the financial services market. At the same time, the technology provides everyone with visibility that the strategy is working.

Large lenders and the people problem
Top lenders already use automation and oftentimes manage multiple vendors for the same technology blended with with custom code and off-the-shelf packages. Despite the heavy investment, these lenders struggle to achieve scale. They either remain focused on the basics or fall into the trap of high-maintenance solutions that need constant management and tinkering to stay effective. Lenders must continually innovate outside the boundaries of their current state.

Connecting the dots between business goals and technology strategy isn’t always in the skillset of the enterprise architects and yet they are frequently relied upon to do the impossible. On their own, these individuals won’t know where to move the goal post for the next round of innovation. It takes a blend of perspectives to make the next move. Start a “Strategy Council” that meets regularly and includes a mix of roles with business visibility and technical know-how.

Adopting and mastering automation can be challenging: Now what?
Successful adoption of AI automation requires a major technical and cultural shift for organizations. Leaders must understand both business and technical strategy to drive alignment and improvement. Developing practice around business problems that connect with technical investments will deliver value over time. Above all, everyone must agree on the problem and share a vision for the north star.