Planning Poker (Scrum Poker)

Planning Poker or Scrum Poker is one of the most important events in Scrum or any other agile development technology. Almost always the team faces the following question:

How do we evaluate this task?

The evaluation of working efforts influences a lot: a number of rating points, the project deadline, and the amount of money the client will have to pay. Each member of a Scrum Team is better at evaluating tasks that fall into his or her field of professional knowledge. When it comes to completing a task, Scrum shifts the accent from personal responsibility to team responsibility. In other words, the whole Scrum Team should be responsible for the evaluation of this or that task. Besides, such an approach allows us to evaluate the time, necessary to complete the task, more precisely, whereas a separate team-member may overestimate this parameter for a number of reasons.

What Planning Poker / Scrum Poker cards are?

There is a large variety of such cards and each team-member can create his own cards. For example, cards, indicating the number of days left for the development.

Here are some of the most popular types of cards.

First popular type of Planning Poker cards:

The set cards represent the The Fibonacci Sequence: 0, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89.

Second popular type of Planning Poker cards:

This type is represented by the following set of cards: 0, ?, 1, 2, 3, 5, 8, 13, 20, 40, 100, «?», «Coffee Cup». The question mark means that a “player” didn’t understand the matter under discussion or doesn’t have enough info to draw a conclusion. The “Coffee cup” card means: “I’m tired, let’s have a break”.

Scrum Poker / Planning Poker process.

One person is the host and he doesn’t take part in the “game”. Different matters, which need to be evaluated, are brought into discussion one after another. At first the “players” are allowed to discuss each matter without evaluation. After this, each team-member chooses a card and lays it face down. As soon as each one has one, the cards are revealed. A perfect result is the one, where there is little or no scatter. But it almost never happens. Among the revealed there will be cards of both the smallest and the largest values. The members, which chose such values are allowed to speak up and explain, why they choose this or that value. This gives the rest of the team more information and ponder after hearing the reasons, or try to explain their own point of view.

Afterwards the cards are played again and usually the scatter is smaller. If it doesn’t happen, the round is repeated once again. It is recommender to set a time-limit for rounds and limit their number. But in most cases, the values are more or less the same after the third round. If there are minor discrepancies, then most attention is paid to the values of a person, directly related to this task.

Typical problems during Planning Poker

As any methodology or technology, Planing Poker has its own set of rules, which prevents the users from making mistakes and makes the implementation of the tool easier.

Anchoring effect in Scrum Poker

The main problem is the anchoring effect. It is a result of the open discussion of the values. Whenever a person, who speaks first says something like: "I think it takes the developers 18 hours to complete this task", others will stick to the same amount of time, and the one who thought the tasks will be finished in 2 days, may think that 18 hours is enough, and the one whose guess was 5 hours, may think that he underestimated the value. From one point of view, it will take less time to reach an agreement, but from the other hand, it will not be effective, but effectiveness it what we do it for. We risk having an opinion of one specific person, rather than the whole team.

Not standing out from the crowd

Another well-known problem appears when the evaluation process is not simultaneous. For instance, one team-member decides that 18-hour term is enough, but due to the two people, whose estimated time was 5 hours, he may think he hastened to draw a conclusion and decide not to stand out from the crowd and change his initial value to the one that is closer to 5 hours.