Planning Poker

Estimation is probably one of the most controversial topics in the Agile domain. The reason is likely related to assumptions about what these estimations are going to be used for. Some fear teams are going to be held accountable for the provided estimates, which will not be driving better estimations since teams are probably going to add some contingency to the estimates. Others claim estimating is pointless in a Complex domain since there are too many unknowns.

The main objective of Agile Estimation is to have better conversations about the problem to be solved. The main job of a Product Owner is to maximize return on investment, and one part of that equation is the size of the work, so estimating is also helping Product Owners do their job.

Popularized by Mike Cohn, Planning Poker is a facilitation technique for estimation. The idea is that the conversation is facilitated so that every team member’s opinion counts, while keeping the process effective. Planning cards or smartphone Apps can be used to share values, typically a Fibonacci like sequence is used: 0 ,1, 2, 3, 5, 8, 13, 20, 40, 100, ?

How to play

At the start of planning poker, each team member is given a deck of cards. For each user story or theme to be estimated;

  • A moderator (PO, SM or Analyst, doesn’t vote) reads the description.
  • Each team member privately chooses a card value based on their gut feeling and understanding of the story description.
  • When everyone has chosen a card, the cards are shown to all.
  • The ones with highest and lowest value provide their arguments for choosing this value.
  • The product owner and or stakeholder answers any questions that the team members have.
  • Each team member again privately selects a card representing his or her estimate.
  • All cards are shown simultaneously so that all participants can see each estimate.
  • Again the ones with highest and lowest value provide their arguments for choosing this value.
  • After the discussion, each team member re-estimates by selecting a card.
  • When there is just one step difference between the remaining cards, the highest value is chosen.
  • If after the third round there is still no consensus on the estimate, the highest value of is chosen.