You can't really, but there is a display trick that may help. Probably the least helpful answer ever!
If you look at the story in the issue view, there's a time-tracking section which can toggle a view of the estimates and work logged for sub-tasks on and off. But it is just a display, and it causes us problems, as there's no simple answer to how it should be working, and I don't think Atlassian really have the right answer for this one.
The problem is that you can place an estimate on all issues - stories and subtasks (This applies to the calculated time remaining, and accumulated work logs etc, but they all follow the same rules, so I'll just say "estimate" from here-on). Each issue has an estimate that applies to that issue only and that becomes complex when trying to work out how sub-tasks might be included.
The reason this is a problem is that different people have different expectations.
Some of us expect to estimate the story at a high level. If we then add sub-tasks with their own estimates, they're separate and don't add to the story estimate to change, we're just splitting the story up into bits. You could expect each sub-task estimate to reduce the size of the story by the estimate because it's been broken out into a separate issue, but then you'll need to handle negative estimates and stories not reporting the overall estimate. Or you could expect each sub-task estimate to increase the size of the estimate because you're adding new pieces of work to it (that would work well in cases where you create a story and can't usefully estimate it until you start to break it up into bits)
Some of us "don't estimate stories, only sub-tasks" and then roll it up, but that breaks when you have stories that you don't need to divide up. Or we could say "only estimate stories" (which is effectively what you have to do if you have JIRA Software), but that leaves you with a bit of a hole for splitting up story estimates into discrete parts.
There's a good chance you'll get a lot of people here who favour one of the several approaches I've skimmed above. None of them are wrong, but none of them are right enough to suit everyone. There are some other less obvious approaches you can take as well. But we're all a bit stuck because there's no single "right" way to do it.
Personally, because I use Agile a lot, I tend to skip estimates on the subtasks completely (and mostly work in story points too), but I also tend to write code (via script runner add-on) to update stories as sub-tasks are edited.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.