Planning Poker Rules

Planning Poker has a simple set of rules, and almost all of the value comes from following them exactly rather than improvising under time pressure. If you want the reasoning behind the technique first, see What Is Planning Poker? - this page is the step-by-step version for running a round.

1. Prepare before the session

Have the stories ready with a title and enough description that nobody is reading it for the first time out loud. Agree on the card deck up front - most teams use a Fibonacci-like sequence (1, 2, 3, 5, 8, 13, 21), plus a card for "I don't know enough to estimate this yet."

2. Introduce the story

The facilitator or product owner reads the story and its acceptance criteria out loud. This is the moment for clarifying questions, not for opinions on size yet - keep "how big is this" separate from "what does this even mean."

3. Discuss, then vote privately

After a short discussion, everyone picks a card that represents their estimate - without showing it to anyone else. This is the core rule of Planning Poker: estimates must be independent, so nobody's number anchors anybody else's.

4. Reveal all cards at the same time

Once everyone has picked, all cards flip simultaneously. In an online tool like Fibopoker this is a single "reveal" action; around a physical table, everyone turns their card face-up together on a count of three.

5. Discuss the highest and lowest estimates

If the votes match (or are close), the estimate is confirmed. If they're spread out, ask whoever picked the highest and lowest card to explain their reasoning first - they usually know something the rest of the team missed, like a hidden dependency or a shortcut.

6. Vote again

After the discussion, vote again on the same story. Estimates usually converge within two rounds. If they still don't, that's a signal the story is underspecified rather than a sign to keep voting - park it, write down the open question, and move on.

7. Record the consensus and move on

Write down the agreed number against the story and move to the next one. Keep a steady pace - most teams timebox each story to a few minutes so a handful of hard cases don't eat the whole session.

Common mistakes

• Letting one person say their number out loud before everyone has voted - this defeats the entire point of the exercise.
• Averaging disagreeing votes instead of discussing them - the discussion is usually more valuable than the number.
• Re-voting more than two or three times on the same story instead of parking it and moving on.
• Estimating stories that haven't been refined enough to discuss sensibly.

FAQ