Doggy Planning
7 CommentsFiled Under: Agile, Estimation, Team Dynamics
Doggy Planning
Posted by Don McGreal on June 20th, 2009 at 1:36 pm
Timing: 10 mins
Ingredients:
- 1 set of planning poker cards per participant. http://www.crisp.se/planningpoker/
Directions:
One at a time, show each of the following dog names and have participants vote by showing an estimate card:
- Chihuahua – this is the smallest dog and should be the reference estimate. Others should be sized relative to it.
- Great Dane – estimates should be very large
- Golden Retriever – estimates should be medium or large
- Poodle – participants should ask for more information, e.g., is it a standard poodle or a toy poodle?
- Newfoundland – this is a lesser known dog; those who don’t know what it is should not vote and should instead ask questions
- Austrian Guildenbaur – this is a trick; the dog does not exist, so no estimates should be given
Learning Points:
- This exercise should get participants familiar with using the planning cards.
- Estimators should not be influenced by others.
- Topics are sized relative to each other, using Nebulous Units of Time.
- People should not vote on topics that are not understood. They should ask for clarification or if it is not in their domain, they should abstain.
Posted by Don McGreal
7 Responses to "Doggy Planning"
Leave a Reply:
Login
Discuss Comments
TastyCupcakes.org is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Powered by WPNow Wordpress Themes.
Hi Folks, I have a few questions about the doggy planning game:
- Do you need to show actual pictures of the dogs or do you just need story cards with the dogs names?
- I guess another learning point is that the game emphasises the relative size aspect of story points
Thanks for making this resource available.
Kind Regards
Russell
Hi Russell,
Thanks for your comment!
No, I do not ever show pictures of the dogs. Although, I just realized that the write-up does not make that clear. I will update it.
It works better when voters are just giving their best guess at the relative size based on their current knowledge. This makes it easier to correlate to software features. I also like it when I get people voting on the poodle based on their initial assumptions (toy vs. standard). The votes expose the assumptions, so we clarify and re-vote. Having pictures of the poodle would make it harder to expose this.
With lesser known dogs, like the Newfoundland, some voters will actually google the dog in order to show others what it looks like before the re-vote. Again, a picture would prevent this.
I totally agree that relative sizing should be a learning point. I don’t know why I missed it. I’ll update that too!
Thanks for the great feedback Russell!
Don
This seems like an unnecessarily complicated derivative of using simple numbers or blocks to show relative size. Unless the entire team is made up of dog-lovers who know the different types of dogs and their relative sizes, why confuse things. I would definitely never introduce a team to planning poker with this model.
If you want to get cutesy, then well-known animals would be better: ant, spider, mouse, dog, horse, elephant.
Thank you for this post!
I’ve translated it in french :
http://www.fabrice-aimetti.fr/dotclear/index.php?post/2011/01/18/Doggy-Planning
Regards,
Fabrice
Javid,
The idea of having unknown animals is to have people thinking about it.
When we start working on a software, we receive a list of requirements that we have never heard about it. So we need to ask and collaborate to get the needed knowledge to work and estimate with more accuracy.
[...] le planning poker par le jeu lors de ma dernière formation j’ai repris le principe du Doggy planning et je l’ai transposé dans le monde des grandes cités de ce [...]
Interesting idea, but I don’t think you have expressed it as clearly as you might have.
On first read I thought the dogs were the planning poker options rather than the stories.