In order to ensure a more intuitive and practical user experience, BSG employed UX principles and practices. The team observed users completing daily tasks and conducted a series of user interviews to determine functionality and usability priorities. The interview process allowed the project team to develop user personas, which were used during the development and testing phases to ensure the system met the requirements of the users. The project team constructed a series of wireframes, which were used to ensure the functionality and look and feel of the system were in line with the users’ expectations.

Once the project team had built an overall picture of the user interface requirements, development began. Two-week development sprints within a Scrum framework were followed. This enabled the team to gain ongoing insight into the system’s functionality, as well as immediate feedback from client stakeholders. This ensured the project stayed on track throughout its duration.

The new system was built to replace a legacy component of a mainframe system, in order to enhance its functionality for the newer debt review and related processes. Packaged as a web application, it included an HTML5 user interface for direct user input (such as data overrides) as well as user reporting. A large percentage of the functionality was developed to perform scheduled batched operations, which needed to be synchronised into the various daily, weekly and monthly data processing cycles of the organisation.

The system therefore needed to integrate with other systems operating in the client environment, such as scheduling and reporting, together with the client’s existing mainframe systems. Furthermore, the system needed to deal with data being received from external sources (e.g. debt review agencies) and needed to be able to handle and report on data abnormalities and exception conditions. It was developed using a combination of Asp.Net MVC, knockout.js, WCF, SSRS and custom Entity Framework—required to handle bulk data ETL processes and be able to process millions of rows within a very short duration.