We've Got Stories Bigger Than Your Iteration.
There are two project teams at one of our clients. Micah Martin and I have been working with different teams for a couple of months now. After a recent iteration meeting we were chatting and Micah commented proudly on how many stories and points they had signed up for that iteration. "Agile Craig" Demyanovich, a team mate of mine said "We've got stories the size of your iteration".
We all had a good laugh over this, but it brings up an interesting point. How do you measure points across teams?
And now, to make things more interesting, we're going to be merging the development teams into one team servicing both projects. I'm really looking forward to the conversations around which project gets more points in a given iteration. And I expect some interesting debate over estimates, given that a point on our team has likely (though we don't know for sure) come to mean something different from a point on the other team.
Feel free to comment if you have any advice. I'll be posting about this as I learn more.
!commentForm
We face a similar issues within our own team... one person's point is not always the same as another's, especially whan a new developer joins the team. We will generally discuss the story and the technicalities behind it and come to a consensus, but the "New Guy" invariable holds back until gaining a feel for how everyone else thinks. It's interesting that our velocities work out in the end fairly close to our estimates, at least better than any other estimating method I've tried. And it's all based on something called a "Point", which no one can really define concretely.
Our team is pretty new at using this methodology for planning. It's been an interesting learning curve with mostly positive results. I look forward to hearing from others who have traveled the same path.
Our team is pretty new at using this methodology for planning. It's been an interesting learning curve with mostly positive results. I look forward to hearing from others who have traveled the same path.
This is a question much debated in the Function Point community, and for exactly the same reason: FPs are an abstract unit. They're built by combining a whole bunch of smaller guesses, but they're still essentially based on the estimators saying the same thing when they see the same thing.
John Roth
John Roth
The International bureau of weights and measures has a bar made of a 90-10 alloy of Platinum and Irridium that has two marks on it. The distance between those two marks is exactly one meter.
One way to solve the problem of points is to create a platinum standard. Post three or four stories on the wall that are the standards for 2-3-5-8 points. (Note the fibonacci series which adds no value to this discussion other than to be "interesting".) When debates about points arise, go to the board and check the estimated story against the standards.
One way to solve the problem of points is to create a platinum standard. Post three or four stories on the wall that are the standards for 2-3-5-8 points. (Note the fibonacci series which adds no value to this discussion other than to be "interesting".) When debates about points arise, go to the board and check the estimated story against the standards.
""New Guy" invariable holds back until gaining a feel for how everyone else thinks"
In response to this, on our team we have recently started to play planning poker. To play you print out cards with the values of your points (1,2,3,5,etc.). Then, during the planning game, while you are estimating, everyone puts down a number face down. That way other developers can't ride off of others understandings of the stories. Otherwise you are estimating to the person who knows the story the best, who is not always the one who implements the story. The poker game provides this veil of ignorance from which the stories end up with a more acurate estimate due to a better understanding and a more independant bet estimate.
In response to this, on our team we have recently started to play planning poker. To play you print out cards with the values of your points (1,2,3,5,etc.). Then, during the planning game, while you are estimating, everyone puts down a number face down. That way other developers can't ride off of others understandings of the stories. Otherwise you are estimating to the person who knows the story the best, who is not always the one who implements the story. The poker game provides this veil of ignorance from which the stories end up with a more acurate estimate due to a better understanding and a more independant bet estimate.
Add Child Page to WeveGotStoriesBiggerThanYourIteration