“I’ve found iteration burndown charts to be extremely useful with iteration lengths of two weeks or more.” ~ Mike Cohn in Agile Estimating and Planning
“I’ve found burndown charts frustrating, irritating and misleading.” ~ Me on a regular basis when they are discussed
Mike Cohn goes on to say “An iteration burndown chart plots hours remaining by day”. Hours? As we keep getting told that humans aren’t good at estimating time, and software development has even more uncertainties than many other situations, calculating how long stories are going to take in hours is something we avoid by using tools such as planning poker. So why in the name of the agile gods would we want to spend a significant amount of time calculating how many hours all the stories in an iteration total? Is a burndown chart useful?
Yes, in theory, as it shows us our progress in the iteration. Are we on target to finish our stories by the end of the iteration? But, as anyone who has used a burndown knows, this rarely works out … because we’re not very good at estimating time. My question is: “what does a burndown tell you that your agile board doesn’t?” Our board tells us where stories are in the process. Our stand-up clarifies how the person working on each story is getting along.
On the other hand, a burndown chart can suggest that the iteration is on target and convince us all that everything is fine, when that is not the case. One story may be going quicker than expected and another going disastrously wrong; but the burndown chart would even itself out and suggest everything is on track. It’s just a dangerous tool if not used in conjunction with the board.
So, if you’re using your board well, why do you need the chart? If you just use the board, won’t you find that more focus is placed on getting updates on the real position of stories and save you a lot of time and hassle worrying about how many hours you think each story will take?