Elaborating user stories by breaking them into tasks

In some teams, the iteration or sprint begins with the team discussing each story and breaking it down into the specific tasks that each team member will need to perform in order to complete the story. The team then track tasks on their story wall rather than complete stories.

Tips

There is a common rule of thumb that no task should be more than 1-3 days work, so that the team can work rapidly in small chunks. This allows the team to have momentum and to focus on what needs to be done next.

Most teams break the whole iteration worth of stories down into tasks on the first day of the iteration. This is a good approach because it allows the team to see where some tasks might be combined for multiple stories.

However I like to have the team start on a subset of the stories to break down first. Then as those are completed or likely to be completed, to grab more stories and break them down. I usually get the developer, tester etc working on the particular story to do this rather than the whole team. However this carries the risk that we will miss integration issues or synergies. So the team will need to decide the best approach in their specific situation.

Some teams also add “unplanned” tasks on a different colour card to record work that was not anticipated in the original planning. This can be useful information for the team to reflect on at the end of the week.

Watch out for

Some teams will leave the task breakdown to one team member and then wait to be told what to do next. This puts a lot of pressure on one person and discourages joint accountability and input.

I like to show “Stories coming soon” to show which stories we haven’t started on. I also like to show them as “100% – we are sure they will be in this iteration”, “80%”, “50%” and “10% likely to make this iteration”.

I find this helps the team start on the most critical stories and push them through before moving to the less critical ones. However I have been warned that this could let the team off the hook in committing to complete the whole lot. It also confuses the team if they are using a burn down chart during the iteration and are not sure whether to include all the stories in the target.