Boosting Financial Platform for a "Big Three" Credit Rating Company

Client

The client is a world leader in the credit rating and research field that provides financial data on companies, countries, and industries.

Client Need

For years, the client had been providing vast datasets to their customers, but as expectations grew and technology advanced, their platform began to show its age. What was once a reliable solution was now struggling to keep up with slow data retrieval, the outdated design, and the lack of advanced features.

To stay ahead and maintain their leadership in the field, the client needed a partner to transform the platform, making it faster, more intuitive, and capable of handling complex data with ease.

Microfrontend setup

To simplify the development process for several websites within the client, the microfrontend system was established. And what is a microfrontend in this context? It’s an approach that enables different teams to develop and deploy product-specific parts of the application independently while maintaining a consistent look and feel across all sites. Our expertise in microfrontends helped minimize duplicated effort, accelerate updates, and ensure seamless scalability as new demands emerged.

Challenges

Chart and library optimization

According to customer requirements, the system had to include some highly specific tables built‌ on a limited data set. However, the capabilities of the databases weren’t sufficient to present them properly. To bridge the gap, we customized chart libraries, fine-tuning features and adding unique elements, transforming raw data into clear and insightful visualizations.

Performance bottlenecks

To boost front-end performance, we turned to Lighthouse — our go-to tool for diagnosing speed issues. It quickly revealed the culprits: oversized bundles, heavy images, and unnecessary rendering, all slowing down page loads, especially on data-dense sections like maps, widgets, and tables.

With these insights, we implemented lazy loading to delay non-essential content rendering. This significantly improved page load times, creating a faster and more seamless user experience.

Nested-checkbox logic

The platform’s nested checkbox logic, used for multi-level selections, often behaved unpredictably, leading to incorrect collapsing or expanding of sections. This made it difficult for users to manage selections and was further complicated by numerous edge cases.

We revamped the nested checkbox logic to be consistent and predictable, eliminating the problems and enhancing the user experience.

Regional permissions

Because the client provides data to users worldwide, they faced the challenge of varying content restrictions in different countries. To ensure compliance with local laws, we needed to geo-restrict certain content depending on the location of the user.

For this, we developed a robust permissions system that dynamically adjusted access rights according to geographical regulations. This ensured users could seamlessly access the information permitted in their region without being affected by restrictions imposed elsewhere.

PDF export

Presenting and exporting large datasets in PDF format was essential for users who needed to save, share, or print reports, but it came with significant technical challenges. Data analysts were publishing reports in HTML, which made saving and printing inconvenient for users. Converting these reports to PDFs often led to formatting issues, making it difficult to produce clear, structured documents.

To solve this, we came up with a solution that transforms HTML research documents directly into PDFs within the browser. By leveraging libraries that work on top of Chrome, we enabled effortless conversion with proper formatting and usability. Now, users can easily save and print reports without hassle.

Excel export

With records dating back to the 1990s, constant updates, and varied storage formats, managing and retrieving information became increasingly complex, especially when users needed to export data to Excel. The challenge was as follows: data came from two sources, formats were inconsistent, and manual processing was slow and inefficient.

To solve this, our team enhanced a Data Lake in AWS. We automated data extraction, transformed it using scripts, and merged everything into a single structured dataset. This enabled fast and accurate downloads in the required format, turning unstructured data into a seamless experience.