William C. Wake's Blog

Using actuals with estimates - an experiment

We also track actuals. Whenever it's convenient, but definitely at the end of each day, the pair adds its tick marks onto the board.

We've used a couple approaches to coming up with "the" estimate. One was to have the longest-standing team member propose a consensus number => not consistently right. Then we tried taking the maximum estimate => usually too high.

Now, we're keeping track of who made what estimate. We compare estimates to actuals and decide who was the "winner" - the one with the most accurate estimates that week. When it's time to do new estimates, we let that person hear all the estimates, and they make the overall estimate we record.

It's too early to say how well this works. The same winner showed up twice in the first month, which may be an early indicator.

My estimates? I'm almost always too optimistic. So I'm going to pay more attention to how the winner thinks about our tasks.

Comments

• Interesting. Did the original XP way not work or is this a way to handle part of it or do you have other restrictions?

  
  The way I remember (although the waterfall-rupies never let me try it) is that the team estimates on stories that are written, the whole iteration is planned from those estimates. Then tasks are created from the stories, each person grabs tasks and estimates them in their own currency and actuals are measured to see the exchange rate between your own hours and an iteration period. I think there is a continuous adjustment of the story estimates as the iteration proceeds. I.e. the purpose is not (intially, but the team will get better) to get the most accurate estimate up front, but to get a more and more accurate prediction as you near the end.

  Anyway, be interesting to read your XPerience :-)

  Posted by: tobega on December 17, 2006 at 10:30 PM

• This is for story-level estimates at iteration tasking. We break the stories into tasks before estimation. We don't break out task-level estimates (we moved away from it a long time ago). (As for XP-style planning, I think most people are working more simply than what the first-edition book described.)

  Posted by: wwake on December 18, 2006 at 05:22 PM

• Hmm, perhaps you could then use the "currency conversion" method. Let everyone estimate, take a weighted average. The weights are each persons ratio of deviation from the actual (and perhaps another weight based on how close your converted estimate usually is to the actual).
  The idea is that even though you are always too optimistic, you may in fact be the best estimator because you get the relative numbers more right than anyone else and your optimism factor is a constant.

  Posted by: tobega on December 22, 2006 at 07:43 AM

This work is licensed under a Creative Commons License.