A lot of people use Jira as the Requirement tool, or rather as the documentation of things to do. This is understandable as Jira is designed as an Agile tool focusing on Kanban and Scrum. Both these methodologies have little to no focus on requirements, as it is implied in the process of iterating thoughts rather than producing defined features. In a simple situation where the team know the product or system and there are little to no external dependencies, this should work just fine. In complex organizations, where there are external dependencies and perhaps even external team members working under contract, this is not a very good way to work. The reason for this is that Jira is not designed for documentation, as it is a work tool with temporary data that should be deleted regularly. Jira also have no version control, even if it has history functionality. This makes it a poor choice for documentation that is legally binding, which is what a requirement is. The second reason we use Confluence is that we want to have a full documentation of systems and products in one place.
