Comments on: Planning Poker Party (The Companion Games)

By: jwgrenning

jwgrenning — Thu, 01 Oct 2009 18:34:04 +0000

Fabrice

thanks for the translation!

James

By: Fabrice Aimetti

Fabrice Aimetti — Thu, 01 Oct 2009 15:59:10 +0000

Hi James, your initial article "Planning Poker or How to avoid analysis paralysis while release planning" is a reference. I've translated it in french : Planning Poker ou Comment éviter la paralysie lors de la planification de version Best regards, Fabrice

By: jwgrenning

jwgrenning — Wed, 23 Sep 2009 12:13:54 +0000

You got it right Hugo. When estimating a big group of stories, its helpful to group by similar effort, then guess a point value for each story column. You could use a planning poker approach to putting an estimate for a column, or just start with 1, then discuss and decide each column after that. The point value assigned to the column represents the effort for each individual story.

Traditional planning poker is works great for smaller numbers of stories. It helps to have a relative effort baseline established. The planning poker party is about getting a big group of stories estimated and establishing that baseline.

By: Hugo Baraúna

Hugo Baraúna — Wed, 23 Sep 2009 03:16:15 +0000

Awesome article! I just have a little doubt: after grouping the stories in columns during the Deal and Slide game, I should do planning poker, ok. My doubt is: should I do a planning poker for each story from each column or should I do a planning poker session to each column?

I think that I should do a planning poker session for each story, and not for each column, it’s weird doing planning poker for a column (or group) of stories, as a column does not have a scope or something that I can measure, as size or complexity.

Am I misunderstanding something here?

By: Agile Projects and Portfolios » Blog Archive » User Story Sizing Game

Agile Projects and Portfolios » Blog Archive » User Story Sizing Game — Mon, 04 May 2009 20:45:46 +0000

[...] Cohn gave me feedback that James Greening has grouped similar ideas in his Low, Medium and High Showdown approach. Thanks for pointing me [...]

By: Ryan Brock

Ryan Brock — Thu, 09 Apr 2009 13:28:37 +0000

We used this in our planning game this week and found this method very helpful. It was much easier to assign points to stories once they were grouped by relative effort. Thanks for the great idea.

By: Kirk Thoughts » Daily post (weekly)

Kirk Thoughts » Daily post (weekly) — Sun, 05 Apr 2009 00:41:50 +0000

[...] Planning Poker Party (The Companion Games) [...]

By: Kirk Thoughts » The Road to Scrum is paved with Lego

Kirk Thoughts » The Road to Scrum is paved with Lego — Thu, 02 Apr 2009 14:36:28 +0000

[...] With 8 people, team estimation stalled at times. Need to learn more techniques to keep flow moving. Hank Roark (@hroark on Twitter) pointed me to James Grenning’s Companion games for planning poker. [...]

By: jwgrenning

jwgrenning — Wed, 11 Mar 2009 18:41:46 +0000

Hello Tiago

The short answer is that anytime you need to baseline your estimates you can host a planning poker party using the needed games. But it sounds like your team could benefit from having a longer backlog. If the backlog is only an iteration or two in length, then it is highly likely that the most important/valuable work is not in the backlog.

A healthy backlog extends well beyond the next iteration and likely into the next release. The length of your backlog is largely dependent on the needs of the business (and maybe engineering) to see what is coming. But if it is made up of only the things that the business happened to think of just in time to get the developers something to do, then it is likely less valuable than a well groomed backlog.

With an appropriate length backlog there should be plenty of stories to use for reference.

I am glad you find them useful.

James

By: Tiago Motta Jorge

Tiago Motta Jorge — Wed, 11 Mar 2009 15:38:48 +0000

I’ve found those companion games just awesome! But I have one question: how should I proceed in the next release planning so that the estimates remain in the same scale? Because if I just start with no stories already estimated, the scale can change, and the mean velocity will tell us nothing. Should we use the stories of the last sprint to help us keep the scale?