Easy-to-understand explanation of how to play agile estimation method planning poker

Even in agile, estimates are a point where problems such as divergence and variation are likely to occur.

In the field of Agile and Scrum, new methods are being considered one after another to make more accurate estimates.

Among them, I would like to explain the estimation method and concept of planning poker, which is often used in agile development.
A client may ask for a point estimate instead of an effort estimate.

Understanding how planning poker calculates points and its benefits will surely help you in the future.

What is Planning Poker | Understanding Story Point Estimation

Planning poker is a planning and estimating technique used by Agile teams after the Product Backlog is created. It helps in estimating the effort required to complete the Product Backlog and the complexity of that work.

Planning poker started with physical cards, hence the gamified name.
These cards, which look like playing cards, are used to estimate story points for each backlog item.

The goal is to provide more accurate estimates and consensus among team members so that the team’s work can be planned more strategically.

overview

If this is your first time hearing a planning poker quote, you might think “that’s a rough estimate!”.
Before we get into the details, let’s consider traditional estimating.

Estimates are guesswork to begin with, so no estimate or plan is 100% accurate.
There are many uncertainties, especially in the early stages of a project.

Agile estimating means acknowledging that uncertainty and providing correct estimates.
In other words, we need to go from 100% accurate estimates of goals to truly correct and useful estimates that include uncertainty.

Reasons for estimating in points instead of man-hours/hours

Estimates in planning poker estimate the complexity and effort of the work rather than estimating man-hours and time.

Therefore , it is possible to standardize estimates because it does not depend on individual skills, etc., and can only be determined by correlation with other work .

Why is planning poker good for estimating

In planning poker, you estimate size, not duration.
The point of planning poker is to use relative estimates rather than absolute estimates.

Estimates are only educated guesses. Use all the knowledge and experience you have at hand to guess its size.

Instead of looking at every new work item individually, compare it to previously completed work items.

For example, is it close to a small one from Project A in the past? Or is it closer to the big one in Project B? Humans are much better at guessing than guessing the actual size of an object.

Estimates for software development are prone to discrepancies

Estimates for software development are more difficult than you might imagine, and there are often cases where the estimates differ greatly between the initial stage of development and the completed stage.

As data supporting this, the “uncertainty cone” graph, which expresses the divergence and variation in estimates in software development, is famous.

The “uncertainty cone” was named and published by Steve McConnell after it was published by Barry Boehm.

The uncertainty cone graph shows a 16-fold divergence from 0.25x to 4.0x in the early stages of development.

As development progresses and decision-making is repeated, the range of deviations in estimates becomes smaller, but the deviations and variations in estimates in the early stages of development have become a major problem, especially in waterfalls .

Planning poker is not about perfect estimates

Planning poker is an effective method to reduce the deviation and variation of estimates, but planning poker is not a method for aiming for a perfect estimate.

As a way of thinking, it is an estimation method to minimize the large discrepancy in estimates that tends to be seen in the initial concept and the early stages of development, which is indicated by the “uncertainty cone”.

It’s not about using planning poker to get a perfect estimate, it ‘s about understanding what problems planning poker can solve.

Next, I will explain what kind of problems can be solved by estimating with planning poker.

What can you solve with an estimate in Planning Poker?

By implementing planning poker, you can solve the following problems.

By estimating with all development members, gaps in estimates are less likely to occur

Even engineers in charge of the same product tend to have discrepancies and gaps in their estimates.

To reduce the likelihood of such gaps in estimates, Planning Poker allows all team members to submit their estimates at the same time.
Also, in advance, there is a description of the product backlog to be estimated and a shared concern.
By taking such steps, gaps between members are less likely to occur.

Planning Poker also avoids the tendency to be swayed by the opinions of superiors and senior members, as all team members estimate.

Planning In his poker, the presence of hierarchies within the team does not affect the estimation process or results.

Faster and more accurate estimates than subjective estimation by one engineer

This is similar to the previous case, but bosses and senior members don’t always know the details of the task.
Many members of the field have knowledge of more specific and technical tasks from different perspectives.

Also, using story her points allows us to think more abstractly instead of using time-based units such as ‘days’, ‘hours/person’. This also reduces the time to reach agreement and brings the results closer together.

Easy-to-understand explanation of how to play planning poker

From here, I will explain how to actually do it.

[Preparation] Creating and preparing planning poker cards

Planning poker used to be played with physical cards, but now there are many tools available. In addition, even without using tools, it is also possible to simply prepare such as dot voting.

How to play planning poker

Here, if the team has different opinions, there is a method of continuing the discussion until the team agrees, but talking about “opinions” may introduce bias and reduce the quality of the estimate. On the other hand, when the team talks about the “facts”, the quality of the estimates improves. For example, if a senior employee says, “This job is easy!” everyone else will lower their estimates, even if they are wrong. To reduce bias, refrain from speaking after sharing facts.

When do you estimate the timing in planning poker?

When is the best time to estimate with Product Parker ?

Planning Poker is conducted immediately after the initial Product Backlog is created.

Product backlog items (user stories) are subject to constant change throughout the project, so estimates are often reviewed at least once per iteration.

Three advantages of planning poker relative estimation

All team members have a voice, so they can work on the project as their own

Planning poker requires the knowledge and input of all participating team members. This ensures that all team members have equal opportunity and responsibility for estimating.

Task estimates are easier to judge

When estimating a task in planning poker, it is judged relative to the user story that serves as a basis for comparison.
Compared to the standard user story, it is easy or difficult to judge numerically, so there is an advantage that it is easier for members to have a common understanding of task weighting and estimation.

Relative estimation improves accuracy

Planning poker considers relative estimates while comparing past results, etc., rather than absolute estimates such as days and hours. Not only does this reduce the time it takes to estimate work, it also improves the accuracy of the estimate. Our brain cannot make absolute estimates. Humans are always thinking about estimates by comparing new things in relation to knowledge and experience they already know.

Planning Poker Q&A | Tell me Joe!

We asked the Agile industry leader, Joe Justice, directly!

Tell us about planning poker from your experience, JOE.

Is estimating valid in Planning Poker even in Agile and Scrum ?

I believe it should be “No Estimate” and I feel it is becoming so.
However, at present, there will be many cases where an estimate is required. In that case, I think the planning poker quotes might help.

What should I pay attention to when estimating with planning poker?

It’s important to share facts and dispel concerns before each person presents planning poker. Discussing opinions after presentation can be biased and inaccurate.

What is the Scrum Master’s job in planning poker?  What should I be aware of?

Have fun doing it!
Be aware that your team will find playing planning poker to be a fun collaboration. If you spend an hour or two talking and estimating, you forget to enjoy your work. There are many ways to add to the fun. You can make original cards with your team members, or you can listen to music while drinking coffee together.

Will a Scrum Master be an advantage when playing planning poker?

In our class, we will practice how to estimate while also talking about “No Estimate”. These will lead the Scrum Master to play with confidence.

summary

Various studies have been conducted on estimation.
Some of them show that planning poker can be more accurate than traditional estimation.

We have explored this possibility in this article. Please try it once!