Loading Profile...
Correctly using 0-point estimates for features?
I've just started using Pivotal. In previous workflow management tools, we used the concept of a "Quick Fix" for something that would take < 1 hour to implement.
Naturally, it seemed that we should choose the "lowest" point score to estimate such stories. The lowest point score is ZERO, so we used that on a bunch of stories.
However, it seems that Pivotal really treats this as ZERO, in that if you have 10-point velocity it will schedule arbitrary numbers of zero-point stories into the current iteration.
Do I have this right? If so, what's the point, since a zero-point story essentially opts you out of the capacity planning part of Pivotal.
Please advise!
Naturally, it seemed that we should choose the "lowest" point score to estimate such stories. The lowest point score is ZERO, so we used that on a bunch of stories.
However, it seems that Pivotal really treats this as ZERO, in that if you have 10-point velocity it will schedule arbitrary numbers of zero-point stories into the current iteration.
Do I have this right? If so, what's the point, since a zero-point story essentially opts you out of the capacity planning part of Pivotal.
Please advise!
I’m confused
4
people have this question
I have this question, too!
Tell me when someone answers.
The more people who ask this question, the more it gets noticed.
The more people who ask this question, the more it gets noticed.
-
Inappropriate?Hey Alan, yeah, you have that right ... 0 point stories do not contribute to velocity. They are useful for tracking tasks that you don't want to forget, but that really don't cost any time to deal with. Small copy changes are a typical example.
The rule of thumb that I use is that 15 minutes or less can be 0 points. Anything more, or any doubt about how something will be implemented, means it has to be at least one point.
As you suggest, you really want to be tracking the true costs of your stories. Therefore we typically use 0 pointers sparingly and in obvious cases. Anything that has a real cost gets real points. -
Inappropriate?JB- Thanks for the reply. From your documentation and FAQ's it would seem that you'd expect people to use a *chore* or *bug* for such a task, since neither of those contribute to velocity.
I am not sure I agree with the way that works for stories. I guess the easy fix is to always use 1-pt or more for stories, but frankly it seems strange to have to train ourselves to do that.
The problem with 0-pt stories is that if you have 20 of them clearly it doesn't take 0 time. Since the whole idea of velocity and points are to figure out if people are good at estimating so that iteration planning works better, then using the "lowest" setting on the point scale shouldn't *turn off* estimation, as it breaks the entire process.
Why don't you just make 0-pt stories count as 0.1 point or something like that? And then people can use chores if they really don't want something to be counted.
Thoughts?
-
Inappropriate?Hey Alan, if the task in question has cutomer-facing value, we classify it as a story. If it's super easy, we'll make it a 0 pointer. As you noted, a bunch of 0 pointers can add up. So we try to use 0 pointers sparingly and appropriately.
But, for example, if there's 10 legitimate zero-pointers in your iteration that could add up to 2 hours of work, don't worry about it. Other estimated stories may come in below their true point value, or maybe they don't. The point of tracker is not accurate, precise estimates on each story, but rather aggregated data that accounts for variations and exceptions. It all comes out in the wash, so to speak.
That said, your team needs to pick a point system that makes sense for them. If Tracker's policies around 0 pointers aren't making sense to you, you can decide not to use them, and have 1 point be the lowest.
I use the 0/1/2/4/8 scale, cut up like this:
0: 15 mins or less
1: 15mins - 2 hours
2: 2-4 hours
4: half day to a day
8: 1 to 2 days
I find this scale provides nice granularity at the low end for easy tasks. Some 2 pointers may be close to ones, but that's ok. And if a few zeros add up, that's ok too.
My 2-pair team recenlty went through a 6 week streak where velocity stayed uncannily close to 30 each week, never going below 28 or above 34. Our work included bugs and chores, of course. We found that the team's true velocity was fairly predictable, which was nice,
1 person says
this answers the question
-
Inappropriate?Thanks for the follow-up. I am definitely willing to give it a try this way, just asking because on its face it's confusing.
I’m confident
-
Inappropriate?I also don't find the 0 points useful - it takes up space in the story card. If it takes such little time as to be estimated as zero, it seems like a waste of time writing and managing a story for it. But this is a fairly trivial thing to be arguing about.
-
Inappropriate?The zero card in planning poker to me is a like a coffee cup or infinity card. Zero represents stories that are too trivial to even waste time talking about, let alone including in the backlog or an iteration.
I don't use a break-down into time units, JB. I prefer to think in terms of abstract complexity units, instead of ideal time, but I know some people do, so I say let everyone do as they please.
I don't see why pivotaltracker feels the need to impose its own point scale approach (especially the use of zero) on everyone. It feels very ham-fisted. Do the preaching on your blog and not in the features, and give us flexibility in the to choose something which is different or at least more mainstream than your in-house approach.
I’m frustrated
