Story Point accounting – I am on the fence on this one folks...or am I?

Firstly, I'd like to say thanks to everyone for the wonderful comments received in regards my writing. Frankly, I am flattered. I'd also like to apologize again for not writing more regularly. It seems like I just never get the time these days with my busy life.

The scenario I want to specifically talk about today is as follows: 

You plan a Story into a sprint, let’s say an 8 point Story, but the team doesn’t get to finish the Story. How should you account for the Story Points in the current and subsequent Sprints? Let’s just say for explanation's sake that the remaining work for the subsequent Sprint is 3 points.

Well, almost all if not all the Agile thought leaders agree that the team should not claim any points in the current Sprint. So even if the team completed 5 points worth of the Story, because it’s incomplete, the Team adds 0 points to their velocity for the current Sprint. I agree with this.

But here’s where it gets interesting. And there are definitely 2 opinions on this.

In the subsequent Sprint, do you claim all the Story Points originally assigned to the Story (8 in this example) or do you just claim 3 points?

Let me first explain why I like the first option.

Claiming all the Story Points in the subsequent Sprint is good because you actually did all the work and so from a pure “value” perspective (Value delivered toward the overall Product), this is the most accurate.

If you choose to only accept 3 points however, your average velocity actually drops lower than your true velocity. Moreover, the reason why you didn’t complete the story in the first place is because it was probably even bigger than the original estimated size –- that’s why you couldn’t finish it in the first Sprint. So if you only claim 3 points in the subsequent Sprint, you’re actually dropping your real velocity even more than just the 5 original points.

The benefit of only planning for and accepting 3 points in the subsequent Sprint is that it is the most realistic representation of the remaining effort. So your planning for the subsequent Sprint will be more accurate with a more predictable outcome I would think. Additionally, it’s a more conservative way of accounting your velocity. If you think about it, we humans are generally optimistic with our estimates in the first place so going the conservative route I think is not such a bad thing.

I posted this question to the Scrum Development forum. And interestingly, folks were divided in their opinions on how to do this. I think it was Ron Jefferies that suggested that Teams try whichever they think is best.

Personally, I think if you look at the problem from a pure Value perspective, then I think the first option is best, i.e., claim all the points in the subsequent Sprint.

If you value ongoing planning and predictability more, then I think the second option is best as you can plan more realistically based on the actual remaining effort.

Let me know what you think.

Happy sprinting

Jack

www.agilebuddy.com 

 

Comments

[Mike Rochleau-Rice]

I am definately on the option 1 side of the fence on this one (counting the full point value in the sprint in which it was completed). If you value predictability, you would do well to count the full 8 points in your average velocity calculations so when you try to determine how fast you can move through your backlog, you're not using an artifically low velocity value.

[Oscar]

I'm also on the option 1 side. Your team definitely worked hard toward the 8 points story and I think its fair for them to be "awarded" the 8 points on their velocity. Also, the team can say that its not fair for you to artificially lower their achieved points.

This actually happened recently to us. Thinking about it, I concluded two things:
1. Developers tend to increase scope of a story on the go, so I need to teach them that the 8 points are their budget for meeting the story acceptance criteria. So planning is essential: "How can we as a team can MEET THE GOAL with the 8 points restriction?"
2. Scope management is essential for meeting sprint goals and release dates.

[jack Milunsky]

By the way, it was also clear that story points are not a reward or punishment and shouldn't be treated as such.

Thanks for your comments so far.

Jack

[Gil Rooke]

I am unhappy. Your previous article says that the benefit of the points is a reward for the developers when they get their estimating right. If this is so, it is NOT a velocity, it is a fixed value reward. Yet, the discussion is all about some concept of velocity.

Whenever you mix two distinct concepts, you get discussions like the one we see above. There is no great philosophy here. It is lack of clarity of concepts. IMHO