Last week I attended a meeting with the group Agile Madison where our topic of discussion was determining the duration for each sprint item. My input on this was that a plan is a strategy for the successful completion of the project and a schedule will tell us when things will be completed. I also reviewed some ways to determine the schedule (or duration) for tasks in a project (Agile or any other type of project management methodology). The classic PMI ones are: historical information, 3-point analysis and Monte Carlo analysis.
The one that was introduced to the group was Planning Poker. Here is how it works.
If you have a project with a group of developers (or testers, or data engineers) each developer is given a stack of cards that have the following on them: ?, 0, 1/2, 1, 2, 3, 5, 8, 13, 20, 40, 100, infinity. The PM (or scrum master) reads off the sprint item and then the developers pull out their cards on how long it should take and then they flip them over at the same time. You may see instances where they all closely agree and other times one person is way off. This will help them to open the channels to discuss why they picked there duration (maybe one of them had a similar task in the past and knew it was a lot more work then it was first predicted).
Why does this work?
It works because it allows you to bring together a group of experts to discuss each task and in many cases you may end up averaging the numbers to come up with your best guess as to what the reality may be.
So the next time you find yourself in a situation like this, think about pulling planning poker out of your PM toolbox.
Ryan
ReplyDeleteAnother technique which incorporated Planning Poker is the Planning Poker Party. http://www.renaissancesoftware.net/blog/archives/36
The planning poker party has participants organize stories into groups of similar effort. It is very useful when there is a large backlog of stories to estimate. Planning poker works better for smaller batches of stories when a baseline is already established.
James