Seperating stories into API and Frontend tasks

[David Carr]

I'm wondering when you have a project that has multiple environments like an API and a frontend website that have different teams. How do you split user stories up?

For example, a user story to create a team requires API endpoints and user interface on the front end.

I would normally create a user story for Create team and then 2 sub-tasks.

User story: Create Team

      Task 1) - API create an endpoint

      Task 2) - front end team create the user interface

Each sub-task has their own time estimates

Curious how other teams approach this.

[Nic Brough -Adaptavist-]

Your approach doesn't work if you're doing anything like Scrum or Kanban.  A sub-task is part of its parent story, and as such, shouldn't be assigned to anyone outside the team that the parent story belongs to.

In your situation, the better practice is to create two stories, one for each team, and link them together to indicate that the teams probably need to talk to each other about it, or that one depends on the other being done.

[David Carr]

thanks that's how we have been doing it, creating separate stories. Looking up options I thought using the subtasks could be a good route.

But I think you're right, the other downside to subtasks is the parent story won't get its estimate from all subtasks.

[Nic Brough -Adaptavist-]

Indeed.  Sub-tasks in Jira really are a fragment of a story.  They're not sprint items themselves either, so estimates on them don't work if you're doing Scrum (or Kanban, but for a different reason)

If you're putting time estimates and work-logs on sub-tasks, you can see an accumulation - when you look at the parent issue, the time panel has a tick-box to swap between "time on this issue" and "time on this issue and its subtasks".  You can also see the sums in a couple of places when you use the Σ fields (the issue navigator is the most obvious one)