Implementing an automated testing infrastructure for a large Fintech company
Case Study
Case Study
Home | Case studies | Implementing an automated testing infrastructure for a large Fintech company
We worked with one of the largest investment firms in the world, who currently manages some of the most prevalent names in the current tech space, and employs over 10,000 people across their global operations.
The client wanted us to develop a subledger system for their financial portfolios.
The primary objective of the project was to improve the quality of the product and make it more resilience to regressions.
From the start, the client’s large-scale system posed a challenge. It contained a wide variety of subsystems and various interactions between them, and was already producing billions of dollars in operations.
However, this was a suboptimal solution, as they were using legacy technology, their QA team had limited automation experience, and they had so many security policies in place that getting access was made a very slow process.
A wide variety of subsystems and various interactions between them. Use of legacy systems leading to inefficiencies and scalability problems.
UI tests also proved to be very unstable, with a pass rate below 50%. Previously, the client was still heavily investing in UI automation testing, and their QA resources were not enough to cover their system’s testing needs.
Improved the client’s Java-based testing infrastructure using Java and Selenium. Continuously expanded the framework over two years. Enhanced reusability between different QA teams.
Divided the project into milestones over two years. The first year’s goal was centered around decreasing average release time from 6 to 3 months, and in our second year, we aimed to further decrease this to a new release each month.
Refactored existing UI tests and created new ones using best practices. Shifted focus toward API testing while limiting UI tests to critical scenarios.
Created and distributed a common QA framework. This framework was continuously developed over the two year span of the project, and we built additional features on top to enhance the QA process
As part of a Scrum methodology, multiple pipelines were configured to execute tests on a nightly basis, and this was done through Azure DevOps.
Successfully transfered knowledge over to the client’s internal team, with both code and experience seeing reusability in different internal projects.
80% of all tests were automated
Two times faster releases
97% pass rate for UI tests
Through our consistent work on this two-year project, we were able to significantly and lastingly improve the quality of our partner’s product, making their system more stable and increasing operational potential, with new versions able to be released with increased confidence.
For a lengthy and complex project like this, there’s plenty of lessons we’ve learned, and we look forward to applying them to future challenges.
Book a free consultation. Schedule a meeting