I am a huge fan of using the physical index cards to represent Agile requirements in the form of user stories. I have in the past used them to very rapidly re-prioritise a requirements list by moving the cards about on a table. Seeing the cards up on a story board and their movement across the columns is a brilliant way of showing progress and focussing the stand-ups. There is also no substitute for the physical cards when estimating (whether you use planning poker or an alternative).
But there is one situation when I have found that the ‘purists’ view of using the physical tokens (the cards) is not as efficient as using whatever tool you use to store and manage your requirements set, in our case TFS.
I tried doing timebox planning by laying out the cards for the rest of the project on a table and asking the team to select those which were required in the upcoming timebox and to arrange them in priority order, into three columns – Must, Should and Could. The problem with this is that, with a large number of cards on a table, it can be confusing, and seemed to generate a lot of discussion as people on one side of the table talked about one story while two other people discussed another. Getting decisions was like pulling teeth. And when we finally did, the resulting notations had to be transferred from the cards themselves into TFS, opening up the possibility of human error.
We have since reverted to what other project managers have been doing for a while – using TFS directly. Having the tool (or an Excel view of it) open on the big screen monitor on the wall, we walk through the list, line by line, deciding on its priority and occasionally re-sizing them, as well as allocating them there and then to the timebox being planned by updating the iteration path field, making the update real-time. If the person in charge of the keyboard types something incorrectly, the rest of the team are quick to shout out the error too. By the end of the session, we have the stories we want all allocated to the correct timebox, with appropriate priorities and sizes. It’s not as much fun as playing with cards, but it works.
I’d be interested to hear your experiences on timebox planning using physical cards