Planning Poker Rules and How to Play - Complete Step-by-Step Guide

Planning poker has become one of the most effective techniques for agile estimation, helping teams arrive at more accurate estimates while building shared understanding. If you're ready to start your first session, this guide covers every rule, step, and best practice you need to know.

What is Planning Poker?

Planning poker, also known as Scrum poker, is a consensus-based estimation technique where agile teams estimate the effort or complexity of user stories. Instead of one person providing estimates, the entire team participates, leveraging collective wisdom for more accurate results.

The technique was popularized by James Grenning in 2002 and later refined by Mike Cohn, who trademarked the term "Planning Poker." According to research from Agile Alliance, teams using planning poker consistently report significant improvements in estimation accuracy compared to traditional methods.

Understanding Planning Poker Card Values

Planning poker cards use a modified Fibonacci sequence:

Standard Card Values:

• 0, 1, 2, 3, 5, 8, 13, 20, 40, 100

Special Cards:

• Question Mark (?): Not enough information to estimate
• Coffee Cup: Time for a break
• Infinity: Story is too large and needs breaking down

Why the Fibonacci Sequence?

The Fibonacci sequence naturally accounts for increasing uncertainty in larger estimates. The gaps between numbers grow as values increase, reflecting our reduced precision when estimating bigger tasks. Distinguishing between 1 and 2 story points makes sense; trying to distinguish between 39 and 40 doesn't.

What Do the Numbers Mean?

The numbers represent relative effort, not absolute time:

• 1-2 points: Very small, well-understood tasks
• 3-5 points: Small to medium, straightforward tasks
• 8-13 points: Larger tasks with complexity or unknowns
• 20+ points: Very large, should be broken down
• 40-100 points: Epic-sized work requiring decomposition

The Planning Poker Rules: Step-by-Step

Here's exactly how to play planning poker:

Step 1: Prepare Your Session

What You Need:

• User stories or tasks to estimate
• Planning poker cards for each participant (physical or digital)
• A facilitator (usually the Scrum Master)
• The development team and Product Owner

Preparation Tips:

• Send stories to estimate 24 hours in advance
• Make sure stories have clear acceptance criteria
• Set a 2-hour maximum for the session
• Have the Product Owner available for questions

Step 2: Select a Story to Estimate

The facilitator picks one user story from the backlog, typically in priority order. Read the story aloud and display it where everyone can see.

Good Practice: Read each story aloud even if people reviewed them beforehand.

Bad Practice: Assuming everyone remembers all the stories.

Step 3: Discuss the Story

The Product Owner briefly explains:

• What the feature or task involves
• Why it's valuable
• Any specific requirements or constraints

Team members then ask clarifying questions.

Discussion Guidelines:

• Keep initial discussions to 2-3 minutes
• Focus on clarifying questions, not implementation details
• Document assumptions
• Product Owner clarifies requirements but doesn't dictate effort

Questions to Ask:

• "What happens if [edge case scenario]?"
• "Are there integrations with existing systems?"
• "What's the definition of done?"
• "Any dependencies we should know about?"

Step 4: Private Selection

After discussion, each team member privately selects a card without seeing others' choices. This prevents anchoring bias, where early estimates influence later ones.

Good Practice: Place your card face-down or select it privately in your digital tool.

Bad Practice: Announcing your estimate before everyone's ready.

Step 5: Simultaneous Reveal

When everyone has selected, the facilitator calls for a simultaneous reveal. All cards flip at the same time—no one sees others' estimates before showing their own.

Step 6: Discuss Differences

This is the most valuable part. When estimates vary significantly, discuss why.

When to Discuss:

• Estimates span more than 2-3 adjacent cards
• Someone played a question mark card
• Wide gap between highest and lowest estimates

Discussion Protocol:

1. Ask the person with the highest estimate to explain their reasoning
2. Ask the person with the lowest estimate to share their perspective
3. Let others contribute insights
4. Limit to 5 minutes maximum

Listen For:

• Different understandings of requirements
• Technical considerations
• Testing or deployment complexities
• Dependencies or risks

Good Example: "I estimated 8 points because we haven't worked with this third-party API before. The documentation looks incomplete, so we'll need time to experiment and handle edge cases."

Bad Example: "I just think it's harder than you estimated."

Step 7: Re-estimate

After discussion, estimate again:

1. Private selection
2. Simultaneous reveal
3. Discuss if needed

Convergence Tips:

• If estimates are within 1-2 adjacent cards, accept the higher estimate
• If estimates still vary widely after 2-3 rounds, the story needs more refinement
• Don't force consensus through exhaustion

Step 8: Reach Consensus

Planning poker aims for consensus, not unanimity.

You've Reached Consensus When:

• Most estimates are the same or within one adjacent card
• Team members with different estimates accept the agreed value
• Everyone's concerns have been heard

When Consensus is Hard to Reach:

• Option 1: Take the higher of the most common estimates (accounts for risk)
• Option 2: Average estimates and round to the nearest Fibonacci number
• Option 3: Mark the story for refinement and move on

Important: The Product Owner doesn't vote. They inform, but the team doing the work provides estimates.

Step 9: Record the Estimate

Record the estimate in your project management tool and move on.

Document:

• Final estimate (story points)
• Key assumptions
• Dependencies or risks
• Follow-up questions

Step 10: Repeat for Remaining Stories

Continue for each story. The facilitator keeps things moving within your time limit.

Pacing Tips:

• Aim for 5-10 stories per hour
• Take 5-minute breaks every 45-60 minutes
• If discussion gets too detailed, create a separate investigation ticket

Key Planning Poker Rules Summary

Here are the essential rules:

1. Everyone estimates: All team members doing the work participate
2. Private selection first: Choose without seeing others' choices
3. Simultaneous reveal: All cards shown at once
4. Discuss extremes: Hear from highest and lowest estimates
5. Re-estimate after discussion: Vote again with new insights
6. Product Owner doesn't vote: They inform, not estimate
7. Time-box discussions: Keep conversations focused
8. Document assumptions: Record important context

Common Planning Poker Mistakes to Avoid

Watch out for these common mistakes:

Mistake 1: Letting One Voice Dominate

Bad Practice: Senior developer speaks first; others defer to their estimate.

Good Practice: Facilitator solicits input from quieter members.

Mistake 2: Turning Estimates into Commitments

Bad Practice: "You estimated 5 points, so this must be done in 2 days."

Good Practice: Story points are relative effort, not time. Velocity is calculated over multiple sprints.

Mistake 3: Estimating Implementation Details

Bad Practice: Spending 20 minutes debating Library A vs Library B.

Good Practice: Estimate the effort. Implementation decisions happen during sprint planning.

Mistake 4: Not Breaking Down Large Stories

Bad Practice: Estimating at 40 points and adding to backlog.

Good Practice: Break stories above 13-20 points into smaller pieces.

Mistake 5: Estimating in Hours Instead of Points

Bad Practice: "This is 3 points because it takes 6 hours."

Good Practice: Use story points as relative effort. A 3-point story is half as complex as a 5-point story, regardless of hours.

Tips for Running Effective Planning Poker Sessions

For Facilitators

Want to improve your facilitation skills? Focus on these:

1. Set the tone: Create a safe space where all estimates are valid
2. Watch the time: Keep discussions focused
3. Ask probing questions: "What makes you think this is more complex?"
4. Prevent groupthink: If everyone agrees too quickly, probe deeper
5. Know when to table discussions: Some stories need offline refinement

For Team Members

1. Come prepared: Read stories beforehand
2. Ask questions: Never estimate what you don't understand
3. Share your reasoning: Explain your thinking, especially outlier estimates
4. Stay engaged: Every estimate matters
5. Be open to changing your mind: Discussion reveals new perspectives

For Product Owners

1. Be available: Don't schedule other meetings during estimation
2. Provide context: Explain the "why" behind each story
3. Clarify, don't dictate: Answer questions without influencing estimates
4. Accept the team's estimates: Trust the people doing the work
5. Use feedback: Many questions signal the story needs refinement

Using Digital Tools for Planning Poker

For remote teams, digital tools are a must. Look for these features:

• Simultaneous reveal of estimates
• Various card decks (Fibonacci, T-shirt sizes, etc.)
• Real-time participation for distributed teams
• Clear view of everyone's estimates
• Easy recording of results

Planning Poker App offers all these features for effective estimation sessions, whether your team is in the same room or across time zones. Check out our comparison of the best tools to find what works for you.

Advanced Planning Poker Techniques

The Bucket System

For large backlogs (50+ stories):

1. Create buckets labeled with story point values (1, 2, 3, 5, 8, 13)
2. Place stories into buckets by relative size
3. Review stories that land in multiple buckets
4. Estimate 50-100 stories in an hour

Silent Estimation

For well-understood work:

1. Display all stories at once
2. Silently assign estimates
3. Only discuss wide variances
4. Much faster for routine work

Affinity Estimation

For new teams learning planning poker:

1. Compare stories to a reference story
2. Group similar-sized stories
3. Assign point values to groups
4. Builds shared understanding quickly

Making Planning Poker Work for Your Team

Planning poker rules promote collaboration, reduce bias, and improve accuracy. But the most important rule? Adapt the technique to your team's needs.

Some teams add extra cards. Others modify time limits based on story complexity. The key is maintaining core principles:

• Engage the whole team
• Prevent anchoring through private selection
• Discuss to build shared understanding
• Reach consensus, not compromise
• Use estimates to plan, not measure individuals

Getting Started

Ready for your first planning poker session? Gather your team, pick your stories, and remember: the goal isn't perfect estimates—it's shared understanding and better planning.

Want to convince your team to try planning poker? Here's how to get buy-in and overcome initial resistance.

Every session will help your team estimate better and stay more aligned. The history of planning poker shows why thousands of agile teams rely on this simple but powerful technique.