What is Planning Poker? A Complete Guide for Beginners
Learn what planning poker is, how it works, and why agile teams use it for accurate story point estimation. A practical beginner's guide with implementation tips.
What is Planning Poker? A Complete Guide for Beginners
If you're new to agile, you've probably heard "planning poker" mentioned in sprint planning sessions. But what is it, and why do teams swear by it?
Planning poker is a collaborative estimation technique that helps agile teams determine effort for user stories or tasks. Unlike traditional estimation methods where one person dictates timelines, it harnesses collective team wisdom for more accurate, consensus-based estimates.
This guide covers everything beginners need to know—from basic concepts to practical tips for your first session.
What is Planning Poker? The Simple Definition
Planning poker (also known as Scrum poker) is a gamified estimation method where team members use numbered cards to vote on story points or effort levels. During estimation, everyone simultaneously reveals their card—ensuring all voices carry equal weight.
Planning poker cards typically follow the Fibonacci sequence: 0, 1, 2, 3, 5, 8, 13, 21, 34. Some teams use T-shirt sizing (XS, S, M, L, XL) or powers of 2. The Fibonacci sequence works well because increasing gaps between numbers reflect growing uncertainty with larger, more complex tasks.
Think of it as a structured conversation about effort and complexity—not predicting exact hours, but building shared understanding of relative effort across your backlog.
Why Do Agile Teams Use Planning Poker?
Prevents Anchoring Bias
When the first person says "this will take 8 hours," others unconsciously adjust their estimates toward that number—even if they thought differently. This cognitive bias skews accuracy.
Planning poker eliminates anchoring by having everyone reveal estimates simultaneously. No single opinion dominates before others form their views.
Encourages Participation from All Team Members
Quieter members or junior developers often hesitate to contradict senior opinions. Planning poker creates a level playing field where a junior's "13" carries the same initial weight as a senior architect's "5."
This diversity surfaces important details others might miss. The junior dev might know about technical debt in a module, while the senior member might be unaware of recent code changes.
Promotes Meaningful Discussion
When estimates differ significantly—one person plays "3" while another plays "13"—it triggers valuable conversation. Teams uncover hidden complexity, dependencies, or simplification opportunities.
These discussions often matter more than the final numbers. They build shared understanding, identify risks early, and align technical approaches.
Improves Estimation Accuracy Over Time
Research shows team-based estimation produces more accurate results than individual estimates. By combining perspectives and discussing outliers, teams arrive at estimates that better match reality.
As teams practice over multiple sprints, they develop shared calibration of point values, making future estimates more consistent.
How Does Planning Poker Work? The Step-by-Step Process
Here's how to run a planning poker session from start to finish:
Step 1: Prepare Your Product Backlog
Ensure your backlog contains well-defined user stories with clear acceptance criteria. Each item should fit within a sprint.
Your product owner should be ready to clarify requirements. Clear user stories lead to accurate estimates.
Step 2: Assemble Your Team
Gather everyone implementing the work—developers, testers, designers, and other contributors.
The product owner provides context but doesn't vote on points since they're not doing implementation.
Step 3: Explain the First User Story
The product owner presents the user story: what needs building, why it's valuable, and the acceptance criteria.
Team members ask clarifying questions. You can't estimate accurately without understanding scope.
Step 4: Individual Consideration
Each member privately considers effort required—technical complexity, challenges, dependencies, testing, and other factors.
No one discusses estimates aloud. This prevents anchoring bias.
Step 5: Simultaneous Reveal
When ready, everyone reveals their card at once—either by placing physical cards face-down then flipping, or clicking simultaneously in online planning poker tools.
This simultaneous reveal keeps judgments independent and unbiased.
Step 6: Discuss Differences
If estimates align, move to the next story. When they differ significantly, it's discussion time.
The facilitator asks those with highest and lowest estimates to explain. The "13" voter might know about technical debt. The "3" voter might have built something similar recently.
These discussions reveal information that changes perspectives.
Step 7: Re-estimate if Needed
After discussion, vote again. Estimates often converge as people incorporate new information.
If estimates still vary widely after 2-3 rounds, break the story into smaller pieces, conduct a technical spike, or use the higher estimate.
Step 8: Record and Move On
Record the consensus estimate in your project management tool, then repeat for the next story.
Most teams estimate 10-20 stories per hour, depending on complexity and discussion needs.
When Should You Use Planning Poker?
Ideal Situations
Sprint Planning: The classic use case. When planning your next sprint, planning poker helps determine how much work the team can commit to.
Initial Backlog Grooming: Work through large numbers of unestimated stories efficiently while building team alignment.
When Team Consensus Matters: For cross-functional teams, the collaborative nature helps everyone feel ownership of estimates.
New or Evolving Teams: Accelerates calibration and helps teams develop shared understanding of point values.
When to Consider Alternatives
Routine Tasks: For work your team has done many times, planning poker overhead might outweigh benefits. Quick discussion or individual estimation may suffice.
Extremely Small or Large Items: Stories that are obviously tiny (1-2 points) or massive (need breaking down) don't benefit from the full process.
Time-Critical Situations: When you need quick estimates for urgent decisions, the thorough discussion might feel slow. But rushing estimates often costs more time later.
Getting Started: Practical Tips for Beginners
Ready to try planning poker? Here's how to set yourself up for success.
Choose Your Card Scale Wisely
The Fibonacci sequence (1, 2, 3, 5, 8, 13, 21) is most popular because increasing gaps reflect growing uncertainty. Some teams prefer:
- Modified Fibonacci: (0, 1, 2, 3, 5, 8, 13, 20, 40, 100) for bigger projects
- T-shirt Sizes: (XS, S, M, L, XL, XXL) for relative sizing
- Powers of 2: (1, 2, 4, 8, 16, 32) for mathematical simplicity
Start with standard Fibonacci and adjust if needed. Consistency matters more than specific numbers.
Establish What Story Points Mean
Story points represent effort, complexity, and uncertainty combined—not time. Your team needs shared understanding of point values.
Many teams use reference stories: "A 1-point story is adding a button. A 5-point story is a new API endpoint with tests. A 13-point story is implementing new authentication."
Build these reference points during your first few sessions and refer back when estimates feel inconsistent.
Keep Sessions Time-Boxed
Set a 2-5 minute limit per story. If the team can't reach consensus, either:
- Go with the higher estimate
- Mark the story for investigation
- Break it into smaller pieces
Progress beats perfection. You can refine estimates later.
Use Online Tools for Remote Teams
Remote teams need digital planning poker tools. Platforms like Planning Poker provide real-time voting, instant results, and estimation history.
Look for tools with anonymous voting, customizable card decks, and easy session sharing.
Don't Skip the Discussion
Numbers are the starting point—conversation is where value lies. When estimates differ, don't just average or pick the highest.
Ask "Why that number?" These discussions uncover risks, dependencies, and simplification opportunities.
Review and Retrospect
After a few sprints, compare estimates versus actual effort. Were 5-point stories consistently faster or slower than expected?
Use this to calibrate future estimates. Planning poker improves with practice.
Start with Small Sessions
Don't estimate your entire backlog at once. Start with 30-60 minutes for your immediate sprint. Extend sessions as your team gets comfortable.
Common Planning Poker Mistakes to Avoid
Watch out for these pitfalls:
Treating Story Points as Time Estimates
Story points aren't hours. Don't convert them (like "1 point = 2 hours"). This defeats relative estimation's purpose and creates false precision.
Points account for complexity, uncertainty, and risk differently than time. A simple task in unfamiliar tech might take the same time as a complex task in familiar territory, but they deserve different point values.
Averaging Instead of Discussing
When you get 3, 5, 5, 8, and 13, don't average to 7. Outliers represent important perspectives. Have the discussion—that's where value lives.
Letting One Person Dominate
If your tech lead explains why every estimate should match their number, you're missing the benefit of team-based estimation.
Facilitators should invite quieter members to share reasoning, especially when their estimates differ.
Estimating Too Far Ahead
The further out work is, the less accurate estimates become. Focus on stories for the next 1-3 sprints. High-level estimates for long-term items are fine, but don't spend time on detailed estimation for work months away.
Forgetting Testing and Review
Estimates should cover all work to meet your definition of done—not just coding. Include testing, code review, documentation, and other activities.
Make it explicit: "This 5 includes tests and code review, right?"
Planning Poker for Different Team Roles
Product Owners
Explain the "what" and "why"—what needs building and why it's valuable. Let the team determine effort without influencing estimates.
Answer questions and clarify requirements, but don't vote unless you're also implementing.
Scrum Masters
As facilitator, keep sessions moving, encourage participation, and help resolve disagreements constructively.
Watch for signs the team needs a break, a story needs breaking down, or someone's perspective isn't being heard.
Developers
Share your technical perspective honestly. If you see complexity others miss, speak up. If you know a shortcut, mention it.
Don't defer to the most senior person—your estimate matters even if you're less experienced. You might implement it.
Testers and QA
Testing is crucial to effort estimation. Speak up about testing complexity, edge cases, and integration requirements.
Your perspective surfaces risks and dependencies that development-only estimates miss.
Moving Forward with Planning Poker
Planning poker is more than estimation—it's a collaborative practice that builds shared understanding. Done well, it improves team communication, knowledge sharing, and collective ownership.
Don't worry about perfection from day one. Start with basics, focus on discussions over numbers, and let your team develop its own rhythm.
Planning poker scales with team maturity. Whether it's your first sprint or hundredth, collaborative estimation continues delivering value.
Ready to start? Tools like Planning Poker make it easy with features for both in-person and remote teams. Create your first session and experience the difference.
Remember: the goal isn't perfect estimates—it's shared understanding, better conversations, and continuous improvement. Planning poker gives you a framework to achieve all three.