Planning Poker And Scrum Poker Everything You Need To Know
PLANNING POKER OR SCRUM POKER EVERYTHING THAT YOU NEED TO KNOW
Hi, this week I will be discussing Planning Poker or Scrum Poker; strategies that use consensus-based estimating. Agile teams from around the world use the planning poker technique to estimate their product backlogs. Scrum Poker can be used with story points, ideal days, or any other estimating unit.
Planning Poker in Scrum brings together multiple expert opinions for the agile estimation of a project. This type of agile planning includes everyone: programmers, testers, database engineers, analysts, user interaction designers, and all other personnel involved in the project. Because these team members represent all disciplines on a software project, they are best suited for the estimation task.
How Does Poker Planning Work
A poker planning, or scrum poker session involves product owners or customers and editors. The session begins with each estimator holding a deck of value-based cards ranging in sequence. We recommend the following: 0, 1, 2, 3, 5, 8, 13, 20, 40 and 100. These values represent the number of story points, ideal days, or other units in which the team will estimate.
The estimators discuss the feature and, if needed, ask the product owner questions. When the estimators have finished their discussion, they each select one card privately to represent his or her estimate. The cards are then revealed simultaneously.
If the estimators select the same value, that number becomes the estimate. If their values differ, the estimators discuss their rational. Those who chose the highest or lowest value should share their reasoning with the group before each estimator selects another estimate card, repeating the process.
The estimators continue the poker planning process until a consensus on the value is achieved. If they cannot agree, the estimators may opt to defer the agile estimating and planning of a particular item, pending additional information.
When should we engage in Planning Poker (Or Scrum Poker)
Most teams will hold a poker planning session shortly after an initial product backlog is written. These sessions, which can take several days, are used to create initial estimates which are useful in scoping or sizing the project.
Because product backlog items – often in user stories form – will continue to be added throughout the project, most teams find it helpful to conduct subsequent agile estimating and planning sessions once per iteration.
They are typically held a few days before the end of the iteration and immediately following a daily standup because the whole team is still together.
Tips for Planning Poker in Scrum
The following tips help teams deal with common challenges in poker planning in Scrum:
- Keep discussions productive: A two-minute sand timer is a helpful tool used to teach teams how to estimate more rapidly. To use the timer, the timer is set when someone in the group starts a round. When the sand runs out, the next round of planning poker cards is played.
- Break out into smaller sessions: It is ideal, whenever possible, to break a larger group down into smaller sub-teams. This is a good option for running sessions when there are many stories to be estimated; often occurring at the start of a new project.
- When to play: Typically, estimating teams need to play planning poker on two different occasions: during the first iterations before the project begins and when new stories are identified during an ongoing iteration.
Top 3 Reasons Why Planning Poker is Great
- It fosters collaboration by engaging the entire team.
- It uses consensus estimates rather than single person estimates.
- Through discussion of each user story, it exposes issues early in the project.
Watching successful teams during poker planning sessions will clearly demonstrate the three top reasons listed. Poker Planning is a great tool with many benefits, but there are other ways to help improve the process even more.
Small steps that most people miss, can be done to improve the process. Knowing these tips will enable the successful coach or trusted team member, to help their team improve. So what are these nuggets of wisdom?
Those who do the work should vote.
Too often, agile teams have everyone vote, regardless of their role in the project. Only those actively involved in the story should vote.
Managers don’t vote.
Managers usually want the work to take less time, so they often vote very low. However, they have more experience than the average team member. By giving a manager veto power over the team consensus in one specific circumstance, they can then ask the team to consider something that may INCREASE the size.
Never allow managers to give low sizes or persuade the team to decrease its size because their opinion carries too much weight. A manager’s views will act as an anchor and drag the size down as they vigorously defend their position.
When there is a tie in the voting between two sizes which are consecutive, just pick the larger size and move on.
Consecutive sizes might be 5 and eight8 if you are using the Fibonacci sequence for sizing (1, 2, 3, 5, 8, 13). An equal split usually takes a long time to resolve, therefore, use the higher number. In my experience, no one ever agrees to a middle ground number and usually choose the top size at the end of the discussion. Using these facts is advantageous and will speed up the process.
Stop implementation discussions before they go too deep.
Teams commonly get very technical with details when discussing a user story. While this is fine to a certain extent, it should be severely limited. Discussions should not take too long. The people involved in the sizing should already have an idea for the simplest, plausible solution and pick a size based on that scenario.
While many believe that more discussion will make the size more accurate, the reality is that is not true. It is done in part, to encourage teams to just make their best guess and move on. However, there a distinct lack of precision so the scale is not granular.
Use an “I need a break” card.
Teams often become so involved in their poker planning sessions that they do not realize when some team members need a break. Having an “I need a break” card can be used by someone on the team to draw everyone’s attention to the need for a break.
Use a timer of some sort to limit discussion.
Using a timer to limit discussions is self-explanatory. Discussions should be restricted to no more than one minute.
If consensus cannot be reached by the end of the third round of voting, pick the largest size and move on.
After two rounds of discussion, further attempts will not yield greater results and wastes time. By choosing the largest size, the team has room to improve and will not be in any danger of running out of time. The biggest problem teams try to avoid is running out of time, so this shortcut should resolve that issue.
Have the person creating the user stories meet with QA and Development leads before playing Planning Poker
User stories should have the answers to the most obvious questions ready. Having answers ready, the team can focus on the size, not gathering information.
Remember the baseline.
Whatever the team picks as a baseline, it needs to be consistent from iteration to iteration. If an ideal day is size one, then all iterations need to use that reference point. If a user story is size one or three, then that needs to remain consistent across iterations.
Sometimes, it is helpful to re-address the baseline and discuss with the team what they believe the sizes truly are. Failure to do this could eventually lead to “size creeping” a term I use when the team mentally changes their baseline over a larger or smaller size over time. This usually happens when the team cannot meet their commitment for several iterations, even though everything looks more or less the same.
Playing Planning Poker should be a fun, collaborative exercise. Too many teams try to grind it out for an hour or two and forget to enjoy their work. There are many ways fun can be injected into the process. I like to play real poker with the sizes to add fun.
To play, every sized story counts as a poker card, and every five stories make a poker hand. Before starting, everyone tries to pick which hand will be the best. This encourages the team to look at the user stories ahead of time, and then try to guess which set will make a straight or four of a kind. Winners can even get a small prize.
ARE YOU WONDERING IF YOU ARE DOING A GOOD JOB AS SCRUM MASTER?DOWNLOAD MY ASSESSMENT THAT WILL PROVIDE YOU A GREAT INSIGHT ON YOUR DAILY ACTIVITIES AS A SCRUM MASTERGet My Assessment Now!
Planning Poker (Or Scrum Poker) for Distributed Teams
“How can you play Planning Poker with teams that are geographically distributed”? There are likely many options to achieve this. These are a few common options:
Option-1: Chat Window (Tried, but leaves loopholes for biased points selection)
Communication mediums like GoTo or WebEx are used when teams are geographically distant. Once the requirements are clear, the product owner can signal via an utter to give points, and every member can use chat window tools to enter their number.
Option-2: Webcam (I’m using this one in the current project – Not great fun, But works!)
When using video conferencing to meet with multiple people (Google+ video a free and popular one), one person runs the meeting. That person says “1-2-3-Go”; after “Go” is said, each team member shows their card on the webcam.
Because the tools were supporting multiple people video chat display the users in a mosaic/tiles based form, it is easy to see all the figures at once. However, on a funny note, people tend to show up cards after watching others. To prevent this, ask every member to hold up their card with the backside facing the webcam. Once host instructs the team to show their card, everyone simultaneously flips their card over.
Option-3: Planning Poker.com (Good. But you can’t change cards)
This website is designed by Mountain Goat Software, the same organization that pioneered poker planning cards. It is a free site that allows the moderator to create the project, add stories and invite people via email to participate in the process.
Any user who clicks on this link sees the story and standard scrum cards. The user does not need the authorization to click on any Scrum or post their thoughts on the story’s complexity.
Option-4: Google Docs
This idea is simple: create a spreadsheet in Google Docs and allow people to type their numbers.
The idea is similar to Google Docs, but MS Office documents also offer live synchronizing and platforms. While usage is limited to the desktop browser, but people also can post their opinions from their smartphones. Users new to OneNote will benefit from this video to learn the program.
Option-6: JIRA plugin
JIRA for Scrum management has a plugin to integrate poker planning within a JIRA UI. More details are available here.
Scrummy is a story point estimation game for scrum teams, developed by the Web Chefs at Four Kitchens to allow easier participation by remote team members and to experiment with some new technologies.
Hope you enjoyed this blog post.