Many Scrum teams meet during each Sprint to estimate the required effort for completing upcoming tasks. Traditionally, teams do this during or just before Sprint Planning.
But estimation meetings have a reputation for being tedious and tiring affairs that distract from ‘real work’, like coding, writing, or designing. What’s more, doing estimation meetings as a remote or distributed team can be difficult.
In this guide, we’ll walk through how you can run a great online planning poker session, whether you’re a co-located, hybrid, or fully remote team!
Planning poker is a task or project estimation method in which teams come together to estimate how large or small various projects are using a set of poker cards.
The process helps teams to ensure everyone is assigned an achievable amount of work for future sprints.
In planning poker, teams review a selection of user stories or tasks from the product backlog. Then each participant individually estimates how much effort that task will involve. Teams do this by playing cards with different values from a customized deck.
Everyone chooses their card anonymously, before they are all revealed simultaneously. This helps teams have an honest conversation about how much work selected tasks or user stories require to complete. What follows is a discussion of everyone’s estimates so teams can get on the same page about the effort required to complete pieces of work.
There are a few reasons why you might want to try planning poker compared to other estimation techniques.
Finally, planning poker estimation works well remotely and there are tools on the market to help teams estimate online.
For example, when you run your estimation meetings online with a tool like Sprint Poker, you even save time because you can streamline or automate many facilitation duties.
There are a few things you’ll need to do before you get going with your planning poker meeting:
As a remote team, you can do much of this preparation work asynchronously, before the meeting begins.
Start by ensuring the backlog items you will estimate are sufficiently detailed for participants to make accurate estimates.
In his paper on planning poker, Bob “Agile Bob” Hartman from Agile for All recommends asking Quality Assurance (QA) and Development leads to review tasks before your planning poker session and encouraging them to add relevant details. It’s hard to make an educated estimate if important context or detail is missing.
When your backlog items are sufficiently detailed, make a list of the ones you want to select for estimation during the meeting. Our Sprint Poker tool lets you import your tasks and stories directly from product backlog in Jira or GitHub before the meeting begins, so they’re ready for the team to estimate.
To complete the preparation of your planning poker session, ask attendees to read through all the tasks you’ve selected before the meeting starts.
There are two common estimation methods for planning poker. You’ll need to decide which one to use before you get started:
Since planning poker practitioners shy away from estimating by time (turns out humans are pretty bad at it), you also need to decide what measure – often referred to as the scale – you will use to estimate a task’s effort. These are the numbers or units you will see on your playing cards.
The default scale for planning poker is the Fibonacci scale, a sequence of numbers in which each is the sum of the two preceding digits – 1, 2, 3, 5, 8, 13, 21, and 34. Using this approach, you would play the “1” card for a task that hardly requires any effort and “34” to indicate an impossible amount of work.
Other common scales include T-shirt sizes (👕 XS, S, M, L, XL), priorities (P1, P2, P3, P4, P5), five fingers (1, 2, 3, 4, 5), and even small and large animals – more on those later, promise. 🐶
Before you get started, talk to your team about the estimation method and the scale you’ll use, along with an explainer of any other confusing Scrum terms that might come up.
Assuming you’ve selected what you’ll estimate in advance, you’re ready to start playing planning poker with your team! Now it’s time to start estimating using poker cards.
Only people with awareness and context of a task should vote. It helps to have more people involved than just those doing the actual work, as insiders' estimates tend to be too optimistic – an effect known as the planning fallacy. At the same time, people with hardly any knowledge of the task or project can't give a helpful estimate or drag out the discussion phase for too long with their questions.
The meeting facilitator should start by reading out the task that’s up for estimation and let participants ask clarifying questions to help form their estimate. In Sprint Poker, participants can indicate they have a question by playing the question mark card. Team members can also enter questions and comments in the discussion thread on the right side of the screen.
When you’ve addressed questions, everyone picks a card with the value that most closely matches their estimated size for the task.
Nobody sees anyone else’s selection until everyone has voted. Hidden voting helps to prevent anchoring bias that may skew estimates. This bias is described in detail by psychologists Amos Tversky and Daniel Kahneman in their book “Thinking, Fast and Slow”:
“[Anchoring bias] occurs when people consider a particular value for an unknown quantity before estimating that quantity. What happens is ... the estimates stay close to the number that people considered—hence the image of an anchor. If you are asked whether Gandhi was more than 114 years old when he died you will end up with a much higher estimate of his age at death than you would if the anchoring question referred to death at 35.”
In the context of planning poker, this means if you went around in a circle and knew the person before you picked an eight, you’d be more likely to choose eight or a number close to it, too.
Once everyone’s cards are on the table, it’s time to start a discussion. With a virtual planning poker tool like Sprint Poker, you can see everyone’s estimates clearly and decide on the need for discussion or subsequent rounds of voting to reach a consensus.
If there’s little variation in everyone’s estimates for a task, you can take the average and move on to the next item. In his original paper, planning poker inventor James Grenning says that simply taking an average saves teams time by cutting the required discussion time for most stories from 10, 20, or 30 minutes to just “a minute or so.”
When you do have significant estimation differences, let the folks with the highest and lowest estimates kick off a discussion by clarifying their thinking and then letting others weigh in.
If the team can’t reach an agreement, hold another voting round to see if the discussion has reduced the variation in everyone’s estimates. When that’s still not the case, Grenning has a straightforward recommendation: “If you can’t get consensus, don’t sweat it. It is only one story out of many. Defer the story, split it, or take the low estimate.”
Estimates reflect the effort a task will likely involve so teams can plan their work for the next sprint. They're not promises about a task's completion time or date, so managers shouldn't treat estimates as deadlines.
Image credit: quickmeme.com
Remote teams can also hold the entire voting and discussion process asynchronously with Sprint Poker. Open up Sprint Poker’s voting for a specific period (e.g. 24 hours) and then do the same for the discussion phase afterwards.
It will take longer to complete the entire process, but it allows everyone to participate without having to wake up at odd hours if your team is spread across difficult time zones.
There’s a lot to keep track of during a planning poker session. You might forget something crucial, like this Scrum Master we spoke to:
“At the end of the meeting, there’s so much going on that I often forget to add the final point estimates back into our tracking tool.”
Run through this checklist at the end of your meeting to make sure you don’t overlook anything important when wrapping up:
When you run your session with Sprint Poker, most of these steps are covered automatically. All participants can reference estimates and discussions in the tool at any time after the meeting, and they instantly receive a meeting summary by email.
Sprint Poker also lets you post the final estimates automatically to Jira or GitHub, so you don’t have to worry about manually copying and pasting after the meeting.
The goal of planning poker is not just getting more accurate task estimates; it should also make the process of getting accurate task estimates enjoyable.
To crank up the fun factor, ask your team to create custom estimation scales and values for your poker cards. From Chihuahuas to Great Danes, strawberries to watermelons, or mopeds to monster trucks, the possibilities are limited only by your imagination.