What is Planning Poker? Complete Guide for Agile Teams (2026)

Planning Poker, also known as Scrum Poker, is a collaborative estimation technique used by agile teams to estimate the effort or complexity of user stories and backlog tasks.

How Does Planning Poker Work?

Planning Poker is based on team consensus and uses cards with Fibonacci sequence values (0, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89) to avoid false precision in estimates.

Step-by-Step Process

1. Product Owner presents the user story: Explains requirements and acceptance criteria.

2. Team asks questions: Developers clarify doubts about functionality, technical dependencies, and scope.

3. Private estimation: Each team member secretly selects a card representing their estimate.

4. Simultaneous reveal: Everyone shows their cards at the same time to avoid anchoring bias.

5. Discussion and consensus: If there are significant differences, those who gave the highest and lowest estimates explain their reasoning.

6. Re-vote: The team votes again until consensus is reached.

Benefits of Planning Poker

For the Development Team

• More accurate estimates: Collective intelligence produces better results than individual estimates.
• Shared knowledge: Everyone learns about different aspects of the work.
• Team commitment: By participating in estimation, the team commits more to the work.

For the Product Owner

• Informed prioritization: Better understands the complexity of each story.
• Transparency: Sees how the team perceives the work.
• Realistic planning: Can create more balanced sprints.

For the Scrum Master

• Facilitates communication: Promotes valuable technical discussions.
• Detects problems early: Discrepancies reveal risks or lack of information.
• Continuous improvement: Allows comparing estimates with actual work.

Common Estimation Scales

Modified Fibonacci (Recommended)

0, 1, 2, 3, 5, 8, 13, 20, 40, 100

This scale reflects that uncertainty increases with complexity.

T-Shirt Sizing

XS, S, M, L, XL, XXL

Useful for high-level estimates or teams new to agile.

Powers of 2

1, 2, 4, 8, 16, 32, 64

Simple mathematical alternative.

Special Cards in Planning Poker

• ? (Question mark): “I don’t have enough information to estimate”
• ∞ (Infinity): “This task is too large, must be split”
• ☕ (Coffee): “I need a break”

Common Mistakes and How to Avoid Them

Mistake 1: Confusing Time with Complexity

❌ Wrong: “This task will take 8 hours, so I vote 8”

✅ Right: “This task has medium complexity compared to our reference stories, I vote 5”

Solution: Use relative story points, not absolute hours.

Mistake 2: Being Influenced by Others

Some people wait to see what others vote before deciding.

Solution: Use a digital tool like GoSprintPlanning where everyone reveals simultaneously.

Mistake 3: Discussing Too Long

Sessions drag on indefinitely seeking the “perfect” estimate.

Solution: Use timeboxing of 5-10 minutes per story. If there’s no consensus after 2-3 rounds, take the average or highest estimate.

Mistake 4: Not Maintaining Reference Stories

The team doesn’t remember what a “5” or “8” meant.

Solution: Document 2-3 reference stories for each common value (3, 5, 8, 13).

Remote vs In-Person Planning Poker

Remote Teams

Advantages:

• Digital tools automatically record estimates
• Harder to be influenced by others
• Allows recording the session for future reference

Challenges:

• Less personal connection
• Requires good internet connection
• Harder to read body language

Recommendation: Use tools like GoSprintPlanning that allow real-time estimation without registration.

In-Person Teams

Advantages:

• Greater connection and team energy
• Richer communication
• Easier to use whiteboards for clarifications

Challenges:

• Need physical cards
• Easier anchoring bias if not revealed simultaneously
• No automatic recording

When NOT to Use Planning Poker

Planning Poker is not suitable for:

• Very small tasks: If all tasks are 1-2 hours, it’s unnecessary overhead
• Very technical work that only one person knows: Consider pair programming first
• Very immature teams: First establish a clear definition of “done”
• Under extreme time pressure: In crisis, prioritize execution over perfect estimation

Best Practices

1. Establish reference stories: Define concrete examples for values 3, 5, 8, and 13.

2. Limit time: Maximum 10 minutes per story. If there’s no consensus, split the story.

3. Focus on the “why”: Discrepancies are more valuable than the final number.

4. Update team velocity: After each sprint, review if estimates were accurate.

5. Allow passing: If someone can’t estimate, that’s okay. Don’t force participation.

6. Use appropriate tools: For remote teams, a real-time tool like GoSprintPlanning eliminates friction.

Planning Poker and Story Points

Story points are the most common unit of measure in Planning Poker:

• Not hours: Represent complexity, effort, and uncertainty combined
• Relative: A 5 is always approximately twice as complex as a 3
• Team-specific: An 8 for your team may be different than for another

Calculating Velocity

Velocity = Sum of story points completed in a sprint

Example:

• Sprint 1: 23 points
• Sprint 2: 27 points
• Sprint 3: 25 points
• Average velocity: 25 points

Use this velocity to plan future sprints.

Alternatives to Planning Poker

Affinity Estimation

The team groups similar stories into size categories.

When to use: Initial backlog grooming of many stories.

#NoEstimates

Approach that suggests splitting everything into tasks of 1 day or less.

When to use: Very mature teams with repetitive work.

Data-Based Historical Estimation

Uses past velocity to project without estimating each story.

When to use: Long-term projects with reliable historical data.

Conclusion

Planning Poker is more than an estimation technique: it’s a communication tool that aligns the team and exposes hidden assumptions. When done well, it produces more accurate estimates and a more cohesive team.

Start Now

Ready to try Planning Poker with your team? Create a free session on GoSprintPlanning - no registration, no downloads, just share a link.

Frequently Asked Questions

How long should a Planning Poker session last? Between 1-2 hours for a typical sprint backlog (10-15 stories). If it goes longer, split the session or improve prior refinement.

Should everyone participate in estimation? The complete development team should estimate. Product Owner and Scrum Master can participate but their vote shouldn’t dominate.

What to do if there’s an expert who always knows better? Use the “Delphi” technique: allow anonymous rounds so everyone can give their opinion without authority influence.

Can we change estimates after sprint planning? Generally not. Estimates are a commitment. If you discover new information, note the difference to learn, but don’t change the original estimate.