I just exited a productive mini-meeting about improving estimation of hours and story points for our off-shore teams. One team in particular is struggling, and we got together to brainstorm some ideas on how to help.
I came up with a couple of suggestions that we’re going to try:
- Get the team to think in relative terms. One of the problems is that when the team comes to the table to estimate stories, they start pointing things randomly and in isolation from one another. In other words, a 3-point story might not match a 3-point story in the previous sprint in complexity, or even with another story in the same sprint. As an exercise, when the team is estimating, we will stop and ask them, “So how does this story relate in complexity to this story over here that you gave the same number of points?” Hopefully it will get them to stop and think about it long enough to reassure us that the points are consistent.
- Get the team to sort user stories by difficulty prior to sprint planning. This will hopefully also increase teamwork. The idea here is that prior to the actual sprint planning, the team will get a list of stories that the product owner is thinking about including in the sprint. Then they can, on their own time, work out the list in order from least complex to most complex to implement. Theoretically, by the time they get to sprint planning, they should already know which stories should be estimated higher than others.
We’re not sure how exactly this will work out, but at this point we’re willing to try anything to improve the team’s ability to plan and estimate.
posted on January 8th, 2009 at 4:32 pm
← Back to articles