Discovery Phase 1: Uncovering the Technical and UX Issues

We kicked off the project by diving deep into the system’s architecture and design. Our approach was two-fold: a technical audit and a design audit.

Technical Audit:

Working closely with the engineering team, we identified core bottlenecks. The root cause of the system’s slow performance and inability to scale was clear: it was built on a monolithic architecture with a single database handling all processes—from transactions to reporting. This setup was the key source of inefficiencies, particularly during high-load times like EOD.

Design Audit:

We also performed a comprehensive design audit to evaluate the system’s usability. The findings were troubling. The user interface (UI) was inconsistent, and complex workflows led to frequent user errors. Form fields were poorly structured, causing frustration and wasted time for users trying to complete tasks like generating reports or entering data. Clearly, the system’s design was as much a problem as its architecture.

Discovery Phase 2: Competitive and Competitors Analysis

To further inform our strategy, we conducted a competitive and competitors analysis. We researched how similar systems in the industry were performing and what best practices they followed, particularly in terms of scalability and user experience. This provided us with benchmarks for database performance, API integration, and UX design.

By comparing these systems, we identified opportunities for improvement, particularly in areas where competitors excelled in task efficiency and scalability. The insights gained from this analysis played a significant role in shaping our architectural and UX recommendations.

Understanding User Roles and User Research

We knew that optimizing the system for scalability and speed wouldn’t be enough—it had to work for the people who used it every day. That’s why we took time to understand the user roles within the system and their specific needs.

User Roles:

We mapped out the different types of users interacting with the system:

• Operations Teams: Responsible for managing EOD reports and monitoring system performance.

• Data Entry Staff: In charge of inputting daily transactions, often dealing with complex forms.

• Management and Analysts: Dependent on accurate, timely reports to make business decisions.

• IT and Development Teams: Responsible for maintaining and scaling the system.

Each user role had unique pain points. For example, operations teams were frustrated by the slow processing of reports, while data entry staff struggled with error-prone forms that often required manual fixes.

User Research:

We conducted interviews and usability tests with representatives from each user group. These sessions gave us invaluable insights into how they interacted with the system, where the biggest hurdles were, and what features they wished the system had.

We synthesized this research into user personas, each representing a key user role. These personas guided our design decisions and ensured that we were creating solutions tailored to the real-world needs of the system’s users.

The Solution: A Complete Overhaul

Armed with insights from our audits, analysis, and user research, we designed a solution that addressed both technical and UX challenges.

1. Microservices Architecture:

• We split the monolithic system into microservices, allowing each function—reporting, transactions, and user management—to operate independently. This drastically improved scalability, as new services could be added without overhauling the entire system.

2. Multiple Databases:

• We moved from a single database to a distributed database architecture. Each microservice had its own dedicated database, optimized for its specific tasks. This change eliminated the database bottleneck and sped up processing times, particularly for EOD operations.

3. API Development:

• We introduced internal and external APIs to facilitate easy integration with third-party tools and services. This not only enabled seamless scalability but also allowed the system to communicate efficiently between microservices.

4. Design System:

• We implemented a design system based on best practices (e.g., Material Design). This system provided a set of standardized UI components, ensuring design consistency across all modules.

• The design system also reduced the workload for the QA team, as reusable components required fewer manual tests.

5. UX Enhancements Based on User Research:

• We restructured forms and workflows based on the user personas we had developed. By improving defaults and adding robust validation tools, we significantly reduced form completion errors.

• Task flows were streamlined, allowing users—especially data entry staff and operations teams—to complete key tasks faster and with fewer mistakes.

KPIs and Measurable Impact

With the new architecture and design system in place, we defined several KPIs to measure the impact of our changes:

• Task Completion Rate (TCR): We aimed to improve TCR by 20%, with a focus on critical workflows like report generation and data entry.

• Form Fulfillment Accuracy: By improving defaults and validation tools, we targeted a 15% reduction in form errors.

• System Performance: EOD processing time was cut by over 40%, thanks to the distributed databases.

The Results

Over the course of 20 months, the system underwent a complete transformation, culminating in the delivery of an MVP. The results were impressive:

• Scalability and Flexibility: The shift to microservices architecture allowed for rapid development and deployment of new features, while the APIs ensured easy integration with external tools.

• Performance Boost: Database optimizations and the move to multiple databases drastically improved EOD report generation times, with some processes running up to 50% faster.

• User Experience: The Task Completion Rate (TCR) increased by 28%, and form fulfillment errors dropped significantly. Users reported feeling more confident and efficient in their daily tasks, thanks to the new UI and optimized workflows.

• Design System Impact: The design system not only improved consistency across services but also reduced the time spent by the QA team on testing UI components, speeding up release cycles.

Conclusion

This project was not just about optimizing a legacy system; it was about reimagining how technology could serve the users behind the scenes. By blending architectural innovation with a deep understanding of user needs, we were able to deliver a system that was faster, more scalable, and more user-friendly than ever before.

The client now has a future-proof platform that can evolve with their business, allowing them to scale, adapt, and grow without hitting the bottlenecks they once faced. And for the users? They now have a tool that helps them work smarter, not harder—exactly as it should be.