Free planning poker for your team
Pick a deck, share the link, and everyone votes in private. When you reveal, the whole team sees the spread at once, so the conversation starts from the disagreement instead of the first number said out loud.
Start a session
Free. No account needed. Teammates join with the link and just their name.
- Fibonacci0, 1, 2, 3, 5, 8, 13, 21 and up, plus ? and a coffee card.
- T-shirt sizesXS to XXL, plus ?. Good for epics and roadmap-level sizing.
- Hours0, 0.5, 1, 2, 4, 8, 16 and up, for task-level planning.
- Powers of two1, 2, 4, 8 and up. Like Fibonacci with steeper steps.
- Linear 1 to 10For scoring things that aren't effort, like value or confidence.

A real session with made-up teammates: votes stay hidden until the facilitator reveals them, then the spread is on the screen.
How a round works
A planning poker round takes a couple of minutes per item once the team has done it a few times.
- 1
Read the item and take questions
The product owner or whoever knows the story reads it aloud. Allow clarifying questions, but not solutions. If nobody can answer the questions, the item isn't ready to estimate.
- 2
Everyone picks a card privately
Votes stay hidden until the reveal. That is the whole point: nobody can anchor on the loudest or most senior number.
- 3
Reveal together
If everyone is within a step of each other, take the higher number or the one most people chose and move on.
- 4
Talk about the gap
When the votes are far apart, ask the highest and the lowest voter to explain. The gap usually comes from different assumptions, and surfacing them is what the exercise is for.
- 5
Vote again, once
Re-vote after the discussion and record the result. If it still splits, the item is too big or too vague: split it or take it back for refinement.
Which deck should you use?
The deck matters less than using the same one consistently, but the choice does shape the conversation.
- Fibonacci suits most teams that track story points. The gaps widen as numbers grow, which mirrors how uncertainty grows with size.
- T-shirt sizes suit early sizing of epics, or a team new to estimating. There's no false precision to argue about.
- Hours is for task-level planning where the work is well understood. It invites debate about the exact number, so keep rounds short.
- Powers of two behaves like Fibonacci with steeper steps, so sizes separate more clearly.
- Linear 1 to 10 is handy for scoring things other than effort, such as confidence or business value.
- A custom deck is available from the app's estimation tab if your team has its own scale.
Running it well
- Vote at the same time, always. If someone says a number first, the rest of the room moves toward it.
- Treat the question mark card as information: it means the team is missing something, not that the work is big.
- Don't average the votes. The discussion is the output, and the number is just the record of it.
- Cap rounds at two. A third round is the team repeating itself, not learning.
- If an item is larger than a few days of work, split it instead of estimating it high.
- Never compare estimates across teams or use them to judge people. Points are a team's private planning tool.
Questions
- Do participants need an account?
- No. Share the link and anyone can join with just a name, from a phone or a laptop. The person who starts the session doesn't need an account either.
- Is it really free?
- Starting and running a session needs no account and costs nothing. If you later want to keep your sessions together in a team, there is a free plan, with paid plans for larger organizations.
- Can we use our own cards?
- Yes. Start a session and choose the custom scale in the deck picker, then list the values you want, in order.
- Does it work for a team in one room?
- Yes. Put the session on a screen and have everyone vote from their own device, which keeps votes private even when you sit next to each other.
Estimate your next backlog item
Pick a deck and you'll be in a live session in a few seconds.
Start planning poker