Geeks With Blogs
Aaron Feng Agile Software Development (XP), Test Driven Development, .NET, etc.....

In my old blog I mentioned that we went from three week iterations down to one week. We were not sure if one week iterations will would for us, but week after week we have seen the fruits. So far, we have been running one week iterations for approximately 12 weeks. Everything came to a halt last week when we could not meet our commitment for the week and all of us worked over the weekend voluntarily. Do one week iterations not work for us anymore? During our retrospective meeting we discussed what went wrong last week. We came up with a hand full of reasons, but in the end, it boiled down to improper estimation and vague tasks.

During a short iteration, we have limited amounts of time to deliver business value in our software, so we want to keep our weekly planning meetings as short as possible. I believe we felt obligated to keep planning to the minimal. Now looking back, we had trouble breaking stories to tasks during last week’s planning meeting. That should have been an indication that we might not have a full understanding of all the stories, which means our estimates are totally off. We were spinning our wheels during the task break down, so we might have rushed in order to get it over with. The natural question to ask is why this happened 12 weeks later? In the previous iterations, we were mostly building on something that is already there or has a good foundation. This past iteration we were starting a totally new area in our application, so more time should have been allocated to setup the foundational prerequisites. The reality is planning needs to take as long as it needs. Rushing the planning will only give an illusion of what can be accomplished.

Despite the bad iteration, I still think one week iterations are working well for us. The upside about the past iteration is that we were able to learn from our mistakes and correct it in a very short time. If our iteration were longer, I feel more time would have gone by before we would have noticed something went wrong. The beauty of development is that we can continuously refine our process to keep us on the right path. I am still pretty tired from the weekend work, so I will keep this post short.

Posted on Monday, June 26, 2006 7:57 PM Agile | Back to top

Comments on this post: Running Out of Steam after 12 Iterations?

# Agile Falling on Your Face
Requesting Gravatar...
Steve, one of our senior developers, writes quite elequently on the post I was going to write: we completly...
Left by Sam Gentile on Jun 27, 2006 8:39 PM

Your comment:
 (will show your gravatar)

Copyright © Aaron Feng | Powered by: