Ouija Board Team Estimating

Tags: agile, scrum, estimating, ouija, games

Recently I came across a really fun, innovative way that teams can collaboratively estimate user stories for a Sprint called Ouija Board.  If you're looking for something to liven up your Planning Poker sessions, I suggest giving it a try:

Set-Up and Rules:

As per Planning Poker, the "game" is played in quick successive rounds to get a team consensus estimate (0,1,2,3,5,8,13,21) of a user story or feature.  Break out one of your old decks of Planning Poker cards (or just write numbers on Post-Its) and set up a preferably small, round table like this:

Ouija Estimating:

Place your stack of user stories to  estimate on a another table or shelf that's nearby so they will be easy to "draw" and get estimated.

As usual, you can choose to pick an easy reference user story that everyone can agree is worth "2" points.  Place this story card next to the "2" card on the table.  To begin estimating, gather your team around the table (this works better with small teams of 4-5 people) and draw a user story from the prepared stack, read it aloud and place it in the centre of the table.  Next, have each team member place a finger on the card and gently begin to move it to the cardinal position that best represents the estimate they agree on:

Deciding Estimates

If a card is moved to a number on the table without hesitation, everyone can remove their fingers and the estimate written on the card and can be left by the number as a reference story or removed to a pile designated for estimated stories.

If there is hesitation and a slight tug-of-war to get a card to an estimate, the Scrum Master or facilitator can pause the activity and ask everyone to engage in brief discussion on the impasse.  As per Planning Poker rules, allow only two, two minute rounds of re-estimating and discussion.  If consensus still can't be reached, move the story into a separate pile to review with the Product Owner.

If the story stays in the centre of the table, the team probably doesn't have enough information to estimate.  The story either needs to be broken down and re-estimated, or the Product Owner needs to provide more information or they may need outside input to get confidence on the technologies they need to use to implement it.  Move the story to a separate pile.

If your timebox allows, return to the stories that were set aside and see if consensus can be reached - otherwise, you should be good-to-go.

Check out the video below that inspired this post for an example of how it's played by a real team (turn on closed captioning to see real-time commentary describing the rules).  What do you think? Would this liven up your team estimating sessions? Do you have different games that you play to generate consensus estimates? Tell me more below...

blog comments powered by Disqus