This week's book giveaway is in the Other Languages forum.
We're giving away four copies of Functional Reactive Programming and have Stephen Blackheath and Anthony Jones on-line!
See this thread for details.
Win a copy of Functional Reactive Programming this week in the Other Languages forum!
  • Post Reply
  • Bookmark Topic Watch Topic
  • New Topic

[ Agile ] What if you fail to meet the iteration deadline ?

 
Pho Tek
Ranch Hand
Posts: 782
Chrome Python Ubuntu
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Hi,

I've just started reading Jim Highsmith's APM book and I have a question to ask: Imagine that we've identifed a project which will take over 30 iterations. What should happen if one of the iterations with an uncertainty factor of "Routine" becomes "fluctuating". In this case, the developer might have underestimated the feature and now, he/she cannot meet the initial estimates for iteration x. What does an agile project manager do ?

Thanks

Pho
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
I haven't read that book yet, but in my opinion the sensible way for the project manager to react is to take that new information and incorporate it into the project plan to make it more accurate. That is, correct the estimate for misestimated task, think about wether that new experience suggests to overthink other points of the plan and replan the features appropriately.

Does that sound reasonable?
 
Pho Tek
Ranch Hand
Posts: 782
Chrome Python Ubuntu
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Ilja,

Yes, in fact re-planning is the most sensible thing to do.

I personally think the possible root cause of this happening is that the developer is new to adaptive methods: in which honest communication is required. Sometimes developers want to impress their peers and can be somewhat unrealistic. A new PM who has not worked with the developer is unable to predict this.

I've been googling around and this PDF article recommends a retrospective meeting at the end of each iteration.
 
Pho Tek
Ranch Hand
Posts: 782
Chrome Python Ubuntu
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
By golly, this website is awesome.

A web-site dedicated to growing the practice of looking back to move forward.
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Originally posted by Pho Tek:
I personally think the possible root cause of this happening is that the developer is new to adaptive methods: in which honest communication is required. Sometimes developers want to impress their peers and can be somewhat unrealistic. A new PM who has not worked with the developer is unable to predict this.


Yes, that's one of the possible reasons. Another is that the developer(s) simply were mistaken with their estimates. "To err is human."

I've been googling around and this PDF article recommends a retrospective meeting at the end of each iteration.


Yes, iteration retrospective are a very powerful tool to improve the process personal practices (or to just adapt them to the current circumstances)!
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Originally posted by Pho Tek:
By golly, this website is awesome.


The book is quite good, too: http://www.coderanch.com/t/93530/Book-Reviews/Project-Retrospectives-Normal-Kerth
 
Stan James
(instanceof Sidekick)
Ranch Hand
Posts: 8791
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Saying there are 30 iterations planned is interesting, too. The more there are, the less certain they are for sure. You can solidly plan one or two and tentatively plan the next couple but I'd guess it's usually not much worth doing any detail on the rest. Saying that there are 30 (or any number) is betting that your velocity will hold true to some expectation. If one iteration comes in under the expectation it might be an isolated technical problem or one poor estimate. If a few in a row come in under expectation, it's time to adjust the scope or the number of iterations.

BTW: You can surely go for too few iterations, too. We had a couple releases scheduled for two iterations, way too short for our imperfect test & feedback loop!
 
  • Post Reply
  • Bookmark Topic Watch Topic
  • New Topic