IBM Japan’s AI agent-driven development support partner, “IBM Bob,” was used to visualise the business logic of an Excel macro and rebuild it as a web application. The project was reported as a case study in which the time required to understand the specifications was reduced by approximately 95% compared with the previous process.
エヌアイシー・パートナーズ used “IBM Bob,” an AI agent-driven development support partner provided by IBM Japan, to visualise the business logic of an Excel macro used to prepare estimates and rebuild it as a web application. The initiative aimed to organise business knowledge that was difficult to understand from the specifications alone, with a focus on the maintainability and continued use of the Excel macro.
IBM Bob’s analysis and visualisation functions were used to organise the structure of the VBA code and its data processing. The person in charge reviewed the generated specifications and documentation, deepening their understanding of the existing specifications and organising the business logic. As a result, the time required to understand the specifications was reduced from approximately 1–3 days to approximately 30 minutes.
As the next step, the organised specifications were used to rebuild the system as a web application. The person in charge reviewed and tested the generated outputs while making improvements, carrying out the process from assessing migration feasibility through testing, performance tuning and documentation. It was confirmed that work previously estimated to take approximately 6 weeks was completed in approximately 2.5 days.
The web application is now being used for business operations. In addition to browser-based access, it enables estimate data to be stored and searched, and past results to be reviewed and reused, establishing a foundation for future feature additions and maintenance. The project is positioned as a practical example of using generative AI to make the understanding and redevelopment of business assets more efficient.