Nisha, just to add my thoughts, I've been down this road literally several times, and while I have been able to do pretty much as Tony had suggested, the problem encountered is usually that there are not enough levels of hierarchy available to do an exhaustive trace. I'm struggling with that again now, where I need
Stakeholder->Need (Epic?)->Feature (Issue?)->Requirement (task?)->Code->Test->deploy
I end up sacrificing something inevitably because I can only go three levels of hierarchy deep (its a bug tracking tool after all, and we've just overloaded it to do things like ALM and Requirements management that many people just don't understand deeply enough to see the weaknesses.
So in my model above, I would end up mapping Features to an Epic, Requirements to an Issue (which then naturally ties into linking version control code check-ins), test cases under the Issue, and that's about as close as I can get. If I bastardize the Feature, I can combine the need there, but its challenging from an architecture and refactoring perspective, as the Need of a stakeholder drove a feature; In a forward trace model, we end up getting myopic about the Feature when we might in fact replace the Feature entirely with a different approach when focusing on the need. From a reverse trace model, the Stakeholder for a Feature is probably someone like a Product Manager, but in fact the Need likely came from elsewhere before PM encountered it, and having trace back to the sources of the need allows for deeper digging and qualification of the need. So you *might* be able to get Product Management to manage their own Need trace, perhaps including Need identifiers into the Epic when it is defined, but bottom line, I haven't seen how to do a full trace effectively in JIRA, Tony's approach gets you there but you need to evaluate what compromises you want to make.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.