Modern Web Application Replaces AS400 System for Financial Aid Processing

A nonprofit organization dedicated to helping state residents plan, prepare, and pay for higher education works with schools and education partners to help students access financial aid opportunities that support their transition from high school to college.

One important part of this mission involves supporting the completion of the Free Application for Federal Student Aid (FAFSA). Completing the FAFSA is a critical step for students seeking financial assistance for college, and schools often work with organizations like this one to identify students who may need support during the application process.

To assist schools in these efforts, the organization developed an internal application. The system connects federal FAFSA submission data with student records from high schools across the state. By matching these datasets, the application can identify which students have submitted FAFSA applications and provide schools with information about their progress.

This visibility allows educators and counselors to help students who may not have completed the process or who need additional support to access financial aid opportunities.

The original system supporting this process had been developed more than two decades earlier and ran on an IBM AS400 platform. Staff accessed the application through a terminal-style interface, similar to a command-line environment. While the system had performed its function reliably for many years, the underlying technology had become difficult to maintain and lacked the flexibility expected of modern applications.

The organization engaged Emergent Software to rebuild the application using modern technologies while preserving the core data-matching functionality that schools relied on.

Emergent Software approached the project by focusing on the core function of the application: accurately matching federal FAFSA records with student records maintained by the state. Because these datasets do not share a universal identifier, the system must compare multiple attributes such as names, addresses, and school information to determine whether records represent the same student.

The legacy system used a custom scoring approach to determine whether records should be automatically matched or flagged for manual review. However, without access to detailed documentation explaining how the original matching logic worked, replicating the legacy algorithm would have required recreating a complex system from limited information.

Instead of attempting to reverse-engineer the original logic, the development team evaluated modern approaches to record matching and entity resolution. This research led to the adoption of a probabilistic record-linking model implemented through the open-source Python library Splink.

Splink uses statistical techniques to analyze similarities between records and determine the likelihood that two entries represent the same individual. By leveraging this model, the team was able to implement a scalable matching process without needing to replicate every rule used by the legacy system.

Alongside the new matching logic, the entire application was rebuilt as a modern web-based platform hosted in Microsoft Azure. The updated system integrates with the organization’s secure environment and allows staff to access the application through a standard web interface using Azure authentication.

The new interface introduces tools such as sortable tables, filtering capabilities, and improved data review features that help staff evaluate potential matches more efficiently. These improvements make it easier to review records that require manual verification while maintaining the accuracy of the matching process.

The new platform provides the client with a modern application capable of supporting its financial aid outreach efforts well into the future. By replacing the legacy AS400 environment with a cloud-based Azure platform, the organization has significantly reduced reliance on specialized legacy infrastructure.

The modernization also improves the maintainability of the system. Instead of depending on a platform supported by a small number of specialists, developers can now work with widely adopted technologies that are easier to maintain and extend.

For staff members who interact with the application, the updated user experience represents a significant improvement. Administrators can log into a secure web interface and review data using familiar tools rather than navigating a terminal-based system.

The probabilistic matching model also provides opportunities for continued refinement. Because the model can be tuned over time, the organization can improve matching accuracy and reduce the number of records that require manual review.

The updated architecture also simplifies infrastructure management. Data processing workflows that previously relied on legacy environments now operate within a modern cloud platform designed to support ongoing enhancements.

Most importantly, the system enables the client to continue supporting schools across the state as they help students complete the FAFSA application and access financial aid opportunities.

Emergent Software continues to collaborate with the client as the platform enters its next phase of refinement and operational use, ensuring the application remains aligned with the long-term goal of helping more students access the financial resources available to them.