Pitfalls and Solutions when Implementing Loan Origination Systems
“The perfect system means nothing if no one uses it.”
Across Africa, too many Loan Origination Systems are overengineered, underutilized, or outright ignored in the field. Why? Because real-world operations rarely match what’s on paper.
After exploring how messaging platforms are transforming financial services and how cultural foundations can strengthen financial inclusion in Africa, I want to dive into another critical aspect of financial operations: implementing effective Loan Origination Systems (LOS). My experiences across various institutions have taught me valuable lessons about what makes these systems succeed or fail in real-world conditions.
The Overengineering Trap
One of the most fascinating patterns I’ve observed is the gap between system design and actual usage. During a visit to a branch in West Africa, I discovered brand new tablets still sitting in their original packaging, tucked away in desk drawers. When I asked about this, the branch manager candidly shared that they were simply too cumbersome to use in the field.
“We tried,” she explained, “but entering all that information while sitting with a client at their market stall just wasn’t practical. So, we collect everything on paper and then enter it back at the branch.”
This wasn’t an isolated case. Throughout my career, I’ve seen this scenario repeat itself across different institutions and regions. What works in theory often stumbles in practice, especially when systems are designed without sufficient understanding of daily operational realities.
The most successful implementations I’ve witnessed start with simplicity at their core. Rather than trying to build the perfect comprehensive system from day one, these institutions focus on essential functionality that addresses immediate operational needs, then thoughtfully expand based on actual usage patterns and feedback from the field.
Controls: Finding the Right Balance
When designing a Loan Origination Systems, there’s a natural tendency to implement every possible control and validation upfront. After all, isn’t it better to prevent all potential problems before they occur? My experience suggests otherwise.
I worked with an institution in West Africa where their LOS required document validation by three different department heads for several steps in the loan origination cycle. What was designed as a robust control mechanism ultimately created such significant delays that staff began circumventing the system entirely. This led to manual workarounds for reporting and persistent frustration among everyone involved – loan officers, managers, and even clients who experienced longer wait times.
The system’s controls, intended to reduce risk, instead created a new operational risk as processes moved outside the system where they couldn’t be properly tracked or managed. This wasn’t because the staff were trying to avoid oversight, but simply because they needed to serve their clients in a timely manner.
Christophe Bretagnolle
Digitalization Expert
With more than 20 years of experience, supporting financial institution with their digitalization projects and strategies, Christophe Bretagnolle is a recognized expert of the banking sector in developing countries, particularly in Africa. His wide knowledge of the solutions and requirements for financial institutions to navigate the evolving and shifting digital landscape give him a unique and critical view of the present and future of information technologies within the development finance sector.
The key insight here is that controls should evolve with your business. A nascent institution with a small portfolio and close supervision can operate efficiently with fewer systematic controls than a mature organization managing thousands of loans across multiple regions. By allowing your LOS to grow with your operation, you ensure that controls enhance rather than impede your work.
Rethinking Field Data Collection
Perhaps no aspect of LOS implementation better illustrates the gap between theoretical design and practical application than field data collection. Watching loan officers balance tablets on their knees, struggling with glare on screens under the bright sun, attempting to navigate complex forms while clients wait – these experiences fundamentally changed my thinking about field technology.
In one particularly telling case, I discovered loan officers were waiting until they returned to the branch to complete their “field” forms on tablets. Why? As one officer put it: “Have you ever tried typing lengthy descriptions on this while balanced on a motorcycle taxi? It’s simply not going to happen.”
What’s fascinating is that the solution doesn’t necessarily require abandoning mobile technology – it just means rethinking how we use it. The most effective approach I’ve seen involves leveraging what mobile devices do best in the field: capturing photos, recording locations, scanning documents, and recording conversations. These data types can be collected quickly and effectively on smartphones (which most staff already own and know how to use), then synchronized with the main LOS for completion back at the branch.
This hybrid approach acknowledges the reality of fieldwork while still leveraging digital efficiency. By designing around how people actually work in the field, institutions can improve both productivity and data quality without forcing staff to adapt to impractical processes.
Designing Around Human Behavior
Before any implementation, I now make it standard practice to observe how staff actually work – not how process documents say they should work. What shortcuts have they developed? What information do they actually use when making decisions? Which steps consume disproportionate time? These insights are invaluable for designing systems that enhance rather than hinder daily operations.
In many institutions, I’ve found that loan officers spend significant time on data entry rather than client engagement. By redesigning workflows and focusing automation on administrative tasks, these institutions have been able to redirect that time toward activities that actually grow the business and improve portfolio quality.
The Future: AI with Human Connection
As we look ahead, AI and automation will undoubtedly transform loan origination systems. Routine tasks like data validation, document processing, and basic underwriting will increasingly be handled by intelligent systems. But my field experience has consistently confirmed one truth: financial services remain fundamentally relational.
People want to talk to people about their money, their dreams, and their businesses. The most successful institutions will use technology to enhance these human connections rather than replace them – handling routine processes automatically so staff can focus on building relationships and making the judgment calls that algorithms cannot.
I envision a future where loan officers spend less time on forms and more time on relationships – where systems handle the administrative burden so humans can do what they do best: connect with other humans.
Finding Your Balance
Implementing an effective LOS isn’t about finding the perfect system – it’s about finding the right balance for your specific context. The balance between control and flexibility. Between comprehensive data and operational efficiency. Between technological sophistication and practical usability.
My experiences across diverse financial institutions have shown me that this balance looks different for each organization. But the principles that guide successful implementations remain consistent: prioritize simplicity, evolve controls organically, design for real-world conditions, focus on integration, and above all, remember that systems exist to serve people – not the other way around.
By embracing these principles, institutions can implement loan origination systems that truly enable rather than hinder their mission of expanding financial access across Africa.
Author: Christophe Bretagnolle
