Planning Poker: a Simple Guide (and a Free Online Tool)
Planning poker is an agile estimation technique that helps Scrum teams size user stories using story points. Everyone votes at the same time to reduce anchoring, then the team discusses differences and converges on a shared estimate.
What is planning poker?
Planning poker (also called Scrum poker or Scrum voting) is a structured way for product teams to estimate the relative effort and complexity of a user story. Instead of guessing hours, the team assigns a story point value.
The key idea is independence: each person selects a value privately, then everyone reveals at once. If votes differ, the team discusses assumptions and risks, updates the story (or splits it), and votes again.
Online planning poker (remote-friendly)
An online planning poker tool makes the process easy for remote and hybrid teams: create a session, share a link, vote in real-time, reveal together, and keep results visible while you refine the story.
With ScrumVote you can start a planning poker session for free and invite teammates instantly. If your team usually searches for scrum poker, the workflow is the same.
How to run a planning poker session (step-by-step)
- Pick the story and read the acceptance criteria together.
- Clarify scope: assumptions, dependencies, edge cases, and constraints.
- Vote privately using a shared scale (commonly Fibonacci: 1, 2, 3, 5, 8, 13, 21).
- Reveal together to avoid anchoring and groupthink.
- Discuss differences (usually highest and lowest voters explain first).
- Revote until you reach a stable consensus story point value.
Want to try it with your team? Start a free planning poker session in under 2 minutes.
Want a deeper walkthrough? Read our step-by-step guide on how to use planning poker.
Planning poker and story points
Story points are a measure of relative size. They combine effort, complexity, uncertainty, and risk. This makes them more stable than hour estimates, especially early in discovery.
How story points support agile estimation
- More consistency: teams compare stories against reference examples.
- Better forecasting: velocity emerges over time and supports sprint planning.
- Less false precision: points avoid pretending you can predict exact hours.
If you want a deeper explanation of relative sizing, see our story point estimation guide.
Planning poker examples for sprint planning sessions
A small UI text update may be a 1 or 2-point story, while an API integration with unknown edge cases may land at 8 points after discussion. The goal is not perfect precision, but shared understanding.
How to use examples without bias
Keep two or three reference stories in your backlog and compare new work against them before voting. This improves consistency across sprints and makes agile estimation faster.
Want a complete planning flow? Continue with Scrum planning and then apply it during your next planning poker session.
Tips to make planning poker work well
- Use examples: keep a couple of reference stories for 1, 3, 8 points.
- Split big stories: if you're debating 13 vs 21, the story may be too large.
- Estimate just enough: aim for sprint-ready refinement, not perfect knowledge.
- Keep it timeboxed: if discussion stalls, capture unknowns and move on.
How planning poker supports scrum planning
Planning poker is often used during backlog refinement or directly inside Scrum planning. It gives the team a fast way to size stories before deciding what realistically fits into the Sprint.
This is especially useful when a team wants to connect story points, velocity, and Sprint goals without turning planning into a long estimation debate.
The Fibonacci scale in planning poker
Most planning poker teams use a Fibonacci-inspired scale: 1, 2, 3, 5, 8, 13, 21, 34, 55, 89. The sequence is not arbitrary. As work grows larger, uncertainty grows faster than linearly — a story that takes twice as long to describe is rarely twice as easy to estimate. The Fibonacci gaps force the team to commit to a size bracket rather than debate the difference between 7 and 8 hours.
The practical effect is that discussion becomes qualitative. "Is this a 5 or an 8?" is really asking whether the team has enough clarity to commit to a moderate story or needs to treat it as something with higher unknown risk. This is exactly the kind of conversation that improves sprint planning.
Why Fibonacci works for story points
- Forces meaningful distinctions: adjacent numbers are far enough apart that choosing one over the other reflects a real difference in perceived complexity.
- Anchors relative sizing: once a team has agreed that a login form is a 3, other stories get sized in relation to it — not in absolute hours.
- Surfaces uncertainty quickly: when half the team votes 5 and half votes 21, the gap signals missing information, not a bad estimator.
Alternative scales teams use
- Powers of 2 (1, 2, 4, 8, 16): simpler progression, preferred by some engineering-heavy teams for its mathematical clarity.
- T-shirt sizes (XS, S, M, L, XL): useful for early backlog discovery when stories aren't yet refined enough for numerical estimates.
- Linear scales (1–10): intuitive but risky — teams tend to over-cluster in the middle and the differences between adjacent numbers feel arbitrary above 5.
Want to try different scales with your team? Open a free planning poker session and choose the scale that fits your workflow.
Planning poker for remote and distributed teams
Remote estimation is where online planning poker tools show their biggest advantage. Physical card decks require everyone in the same room; a browser-based tool removes that constraint entirely. Every team member sees the same backlog item, votes on the same screen, and reveals results simultaneously — regardless of time zone.
Setting up a remote planning poker session
- Share the session link before the meeting: team members can review backlog items in advance, which speeds up discussion and reduces meeting time.
- Use video for the discussion phase: revealing votes is fast; the value comes from talking through diverging estimates. Faces matter for reading hesitation and confidence.
- Timebox ruthlessly: remote discussions lose energy faster than in-person ones. Keep each story to a firm 3-minute discussion cap and capture unknowns in the ticket rather than debating them live.
Async estimation for fully distributed teams
Some teams across multiple time zones run async estimation rounds: team members vote on backlog items asynchronously within a set window (for example, 24 hours), then the facilitator reviews outliers and schedules a short sync call only for stories where the range is too wide for consensus.
This approach sacrifices some of the real-time discussion benefit but dramatically reduces meeting overhead. It works best for teams with stable, well-refined backlogs where most stories are straightforward.
Planning poker vs. other agile estimation methods
Planning poker is not the only way to estimate. Teams sometimes weigh it against three-point estimation, affinity mapping, or bucket sizing. Each has legitimate use cases.
Planning poker vs. t-shirt sizing
T-shirt sizing (XS through XL) is faster for early-stage roadmap prioritisation, where stories are large and fuzzy. It produces rough relative rankings but not the granular story-point values needed for sprint capacity planning.
Planning poker is better once stories are refined enough to discuss specifics. The simultaneous vote-reveal and discussion cycle extracts tacit knowledge that simple size-labelling skips.
Planning poker vs. three-point estimation
Three-point estimation (optimistic, most likely, pessimistic) generates a weighted average effort in hours. It works well for individual tasks with known patterns but creates an illusion of precision that planning poker deliberately avoids.
Planning poker's strength is in surfacing disagreement. When votes differ by more than one or two scale values, the team has discovered a risk or assumption that three-point averaging would have buried in a formula.
Related planning poker resources
Planning poker FAQ
What is planning poker?
Planning poker is a collaborative agile estimation technique where the team assigns story points to a user story. Each person votes independently and the votes are revealed together to reduce anchoring bias.
Why is it called "planning poker"?
Teams originally used physical cards with values like a deck of cards. The "poker" part refers to revealing votes at the same time.
What planning poker scale should we use?
Fibonacci (1, 2, 3, 5, 8, 13, 21) is most common because uncertainty grows with size. The scale supports relative sizing with story points.
Is planning poker good for remote teams?
Yes. Online planning poker tools let distributed teams vote in real time, reveal results simultaneously, and discuss differences to reach a shared estimate.
How long should planning poker take?
A typical vote for a well-refined story takes 1–3 minutes. Timebox discussion, split overly large stories, and capture unknowns to keep sessions moving.
How many people should participate in planning poker?
Include everyone who will do the work: developers, designers, and QA. More than 8–10 people starts to slow discussion without proportionally improving estimates. If the team is larger, split into sub-teams or use representative voting.
What does it mean when everyone votes the same?
Unanimous votes usually mean the story is well understood and well refined. Accept the estimate and move on. Occasionally unanimous high votes signal a story that everyone finds intimidating — a good trigger to discuss splitting it.
Should the Product Owner vote in planning poker?
Most teams ask the Product Owner to observe without voting during the initial round to avoid anchoring the team toward the business priority rather than the technical complexity. They can clarify scope questions before voting begins.
How do you play planning poker step by step?
The Product Owner reads a user story and the team asks clarifying questions. Every member then privately selects a card representing their estimate in story points, and all cards are revealed at the same time. If estimates differ widely, the highest and lowest voters explain their reasoning and the team re-votes until it converges. ScrumVote runs this entire cycle online in real time.
Is online planning poker free?
Yes. ScrumVote is a free online planning poker tool — you can create a session and invite your team without signing up or paying. It works for co-located and remote teams, reveals votes simultaneously to prevent anchoring, and supports the Fibonacci scale and other common story point sequences.