What is T-Shirt Sizing Agile Estimation Technique, its Pros and Cons, and Best Alternative?T-shirt sizing is a relative estimation technique where t-shirt sizes, such as XS, S, M, L, and XL, are used to estimate product backlog items in a seamless way.
Most agile teams do estimates, as it helps teams to understand, discuss, and plan what they are going to work on next. Estimation also helps in breaking a complex project into smaller parts and fitting them effectively in the next sprint. Moreover, a record of estimates also helps to measure the team's velocity, which can be used to measure the team's productivity and create better release plans.
There are many estimates technique available today for agile teams. All of them come with their own set of benefits. In this article, we are going to discuss one of the popular techniques, i.e., T-Shirt Sizing. You will learn how it works, its pros and cons, and its better alternative. So, let's begin our discussion.
T-shirt sizing is a relative estimation technique in which an agile team estimates user stories based on t-shirt sizes, such as XS, S, M, L, XL, and XXL. The term "t-shirt sizing" is originated from the way t-shirt sizes are mentioned in the U.S. Instead of giving them a number 2,3, and 4, they are given sizes, extra small (XS), small (s), medium (m), and so on.
In the T-shirt size estimate, the development team is asked to evaluate a story and give it one of the t-shirt sizes. Instead of using numeric numbers, a team is made to evaluate more dynamically the efforts required by the stories. Once the estimation is completed, the team can even assign numeric values to them.
The concept of t-shirt sizing is based on binning/bucketing where the items with similar sizes are grouped together. The decision of the size is made in a mutual collaborative discussion session. The team sets the size of the user story relative to others. For example, a medium "M" size could mean the item is twice as small as the large "L" item, or four times as small. So, it depends on the team how it wants to consider relative sizing. However, the relative sizes should remain consistent for more accurate estimates.
In short, the T-shirt sizing estimation technique offers a seamless way of estimating a big backlog of large items and getting a rough idea of the total backlog size.
The general steps that a team undergoes during a T-shirt size estimation session are as follow:
Make all members join the session and give them XS, S, M, L, and XL cards.
Product owner briefs about the story to be estimated. The team listens and then asks questions to clarify their understandings of the story. The questions can be like:
Does this story require learning new things before beginning the design stage?
Can this story be completed using the existing code class library or there is a need to write a new one?
Does this story require any special measures for unit testing?
What are the major user expectations targeted in this story?
Once all the concerns are clarified, every member assigns the story a t-shirt size.
- All members will show their cards simultaneously.
If everyone has picked the same card, then that is considered the final estimate. Otherwise, the team conducts the discussion again until everyone agrees on one size.
The team keeps doing step 1 to step 5 until all the stories are estimated.
Once estimation is completed, the team can then estimate the time required to complete them (optional step).
This way, a team can easily complete the whole estimation process in a relative and quick way without stressing much about the time parameter. Afterward, the team can start the design and development process and prioritize things accordingly.
Every estimation technique comes with both benefits and disadvantages. Here are the main pros and cons associated with the T-shirt sizing estimation technique:
- It is a simplified, informal estimation approach, so it empowers teams to quickly estimate a large number of items.
- It takes teams out of their analytical thoughts and makes estimation a fun and creative process.
- It helps teams to understand what they have to do next, what are their priorities, and how much effort they have to input without projecting any strict deadlines.
- In case a development team faces delays or challenges in completing one item, developers can readjust their priorities or extend delivery time easily because they are not committed to strict deadlines.
- It requires consistent relative sizes.
- Estimating based on sizes instead of numeric values is less accurate. Moreover, it also requires converting sizes to numeric values later on to better track effort over time and team velocity.
- There is no mathematical relationship between extra small, medium, and other sizes, so it can sometimes become difficult to compare stories with each other.
- Since estimates are rough without any numeric values, so the team cannot be held accountable if things don't go as projected.
T-shirt size estimation is an easy estimation technique for teams who are switching to agile. However, the major disadvantage with T-shirt size estimation is that it does not involve a numeric scale, so the estimates are rough and less accurate. Therefore, it is difficult to project the team's velocity and turnaround time.
Async Poker for Jira
The development team can start with T-shirt sizing for the beginning, but later it should shift to a more accurate and numeric-based estimation technique. This is where Planning Poker comes into action.
Planning Poker is a consensus-based relative estimation technique that is similar to T-shirt sizing but involves numeric scale mostly based on modified Fibonacci sequence (1, 2, 3, 5, 8, 13, 20, 40, and 100). The general steps involved in Planning Poker are as follow:
- The team gathers for an estimation session and is handed poker-style cards containing values that members will assign later to the stories.
- The product owner narrates the story and the team discusses it, just like in the case of T-Shirt Size estimation.
- Once the discussion is completed, each member gives value to that story.
- If everyone gives the same value, then that is the team's estimate. Otherwise, members with the lowest and highest values are given a chance to present their viewpoints. Afterward, the team discusses further until it reaches a consensus.
Sync Poker for monday.com
Overall, the Planning Poker estimation process is similar to T-shirt size estimation with the only difference of numeric scale instead of t-shirt sizes. So, what makes Planning Poker a better alternative compared to T-Shirt Size? Below of some of the key benefits associated with the Planning Poker estimation technique:
- It involves numeric-based estimation, so it gives a better sense of the complexity of the items and the effort required from the team.
- It removes the concerns of consistent relative sizing. Teams can pick modified Fibonacci sequence or any other number combination to estimate using planning poker.
- It results in more accurate estimates, thereby helping in the better projection of turnaround time.
- Team members become more committed to work, as they have a better visual of the error demanded by the items.
- Previous estimates can be used as a reference when doing estimates of similar items for more accuracy.
- It is one of the most widely used estimation techniques by agile teams.
In short, Planning Poker is a more accurate way of doing estimates. Moreover, it can also be executed remotely to assist team members who are working from remote locations, from different time zones, or cannot attend the physical meeting session. The remote way of doing Planning Poker is known as Asynchronous Planning Poker or Async Poker. In this approach, members are sent the stories to be estimated by email. They read and understand the story at their own pace and send their estimates before the deadline. Afterward, the moderator sees if the final team's estimate is reached. If not, then a quick online discussion session is held to reach a consensus.
Async Poker for monday.com
There are many estimation techniques available for agile teams today to estimate their backlogs. T-shirt sizing is one of the easiest ways to start doing estimates in a relative way. So, if there is a huge backlog of large items or the team is new to the agile framework, then it is recommended to practice T-Shirt sizing for some time. Afterward, the team should move to more accurate and numeric techniques, one of which is Planning Poker. In a nutshell, estimation is important to understand product backlog items and have a better sense of required efforts. So, pick the estimation technique that best fits your corporate culture and start doing estimates.