In Scrum, there is the process of backlog grooming, which, in-part, deals with the breakdown of epics/big user stories into smaller stories that are easier to estimate and consume within an individual sprint.
In JIRA Agile, I am looking for the proper way to mirror the grooming process and still have a manageable list of backlog items and an accurate tracking of estimation of the stories.
Here's an example of the problem I see:
- An epic is created as an individual ticket. (Ticket total: 1)
- That epic is broken-down into (let's say) 3 user stories: (Ticket total: 4)
- We try to estimate the first user story and realized it's too big, so we break down this user story into 2 smaller user stories (Ticket total: 6)
I now have a backlog of 6 tickets that are to be prioritized and managed, when in reality I should only need to be prioritizing the smallest user stories against each other. Also, I may have put a large estimate on a large user story, and then refined the estimates through estimating the sub user stories. (i.e. Step 3 might have a large user story estimate of 20 points, but the sub stories might total to 13 + 5 = 18)
Am I splitting stories the correct way as it was intended in JIRA?
Should I remove the estimates of the larger user stories once I've sub-divided the story, and only focus on the estimates for the smallest stories to prevent double-counting?
- How can I manage user-stories that have ultimately become epics in themselves (and still associate them to the higher-level epic)?
(I've been using the Structure plugin, but it doesn't help me with managing backlog prioritization in the agile board.)