Comments on: Tweaking our Backlog Review Meeting http://www.kfurlong.com/2010/04/01/tweaking-our-backlog-review-meeting-2/ A blog about agile development, quality management, business, and technology Wed, 16 Feb 2011 21:42:55 +0000 hourly 1 http://wordpress.org/?v=3.1.3 By: Siddharta http://www.kfurlong.com/2010/04/01/tweaking-our-backlog-review-meeting-2/comment-page-1/#comment-86 Siddharta Wed, 05 May 2010 21:09:09 +0000 http://www.kfurlong.com/?p=139#comment-86 Variable team capacity during a sprint is a big problem with sprint planning. It essentially invalidates the use of velocity for sprint planning. We had this problem too, which is when we dropped sprint planning and transitioned into a pure pull system. The same system worked for prioritization too. With a pull system, the only call that needs to be made it Now, Later or Never. You don't really need to stack rank everything. So at the time of pulling, you can calculate the estimate with the most up to date data, and then decide whether this is a Now, or Later, or Never. That essentially eliminates the need to groom the backlog estimates and keep them up to date. Variable team capacity during a sprint is a big problem with sprint planning. It essentially invalidates the use of velocity for sprint planning. We had this problem too, which is when we dropped sprint planning and transitioned into a pure pull system.

The same system worked for prioritization too. With a pull system, the only call that needs to be made it Now, Later or Never. You don’t really need to stack rank everything. So at the time of pulling, you can calculate the estimate with the most up to date data, and then decide whether this is a Now, or Later, or Never. That essentially eliminates the need to groom the backlog estimates and keep them up to date.

]]>