The beauty of Agile is that it is not prescriptive. Once organizations have understood the crux of Agile, they can tailor it to suit their needs and define processes that will ensure maximum output. Agile estimation is one such example. These simple, yet effective, techniques set the tone for the entire Agile project and help teams navigate complex projects easily.
T shirt sizing in agile is a relative estimation technique that helps for long term effective planning. In this blog we attempt to deep dive into T-shirt sizing estimation and try to understand its benefits and drawbacks.
Know more about agile vs traditional project management.
Agile often starts with a high-level estimate or a macro view of the project. This helps teams and stakeholders come to a long-term plan for the project. One of the ways that Agile does this high-level estimate is though T-shirt sizing agile, in other words estimating story points using a relative estimation technique.
T-shirt sizing, as mentioned before, is a relative project estimation technique to estimate what a project will need in terms of time, budget, and energy. It is a good technique for teams that are new to agile and want to perform a relative estimation of the project. T-shirt size planning can be further broken down into story points for sprint planning. Story points can be broken down further into hours for sprint execution. While t-shirt sizing is great for release planning and defining project roadmap, story point estimates are more accurate and better for sprint planning.
What is a story point? A Story point is a unit of measure for expressing an estimate for the overall effort needed to complete a particular user story, sprint, or product backlog item. While in traditional project management methods the effort is conveyed in a time format like days, weeks or months, Agile uses story points to provide estimates and these can be provided by considering the amount of work, the complexity of work and associated risks. This is where story points differ from estimating in person-hours which may not consider the complexity or risk that may delay the task. Also, story point estimation is more flexible and is perfect for high level estimation.
Choosing a t-shirt when you walk into a store is simple. You have clearly labelled sizes like S, M, L, XL and all you need to do is pick one. While this may be a relative sizing and each T-shirt size can fit a range of shoulder sizes, it is much easier than having numerical t-shirt sizes like 38, 40, 42 etc.
Similarly, for agile projects, teams can classify items or user stories as extra-small, small, medium, large, extra-large, or double extra-large. This T-shirt sizing estimation eliminates the numerical score associated with story points and helps developers to be more flexible and dynamic about the effort associated with a story.
Depending on the size of the task, the developers will assign a t-shirt size. It’s important that all developers arrive at a consensus on the T-shirt size assigned to each task. The stories are all placed in S, M, L, XL buckets and the time taken to complete all the tasks in the buckets is estimated. Teams can get a relative understanding of how big or small the story is.
The problem with T-shirt's sizing is that it is a relative estimate, so teams would only get a ballpark figure or a range and not an exact estimate of effort needed to complete a user story.
Also, since product backlog increments or PBIs are indicated in terms of T-shirts sizes, it would be difficult to estimate or review how much work is done. Like for example, your report may show that for this sprint you have completed 2 small, 4 medium and 3 large PBIs. This may make it difficult to measure the velocity of work.
The t-shirt sizing is a great way of providing initial estimates and can be used as a first round of estimating, providing stakeholders and the team with a relative or broad idea of time and effort required for the project.
As mentioned above, it is often the first round of planning and starts with the project being split into high level epics which may be given t-shirt sizes. You may give an estimate range for epic size in a number of days.
You can use this estimate and suppose that your first delivery of the product will take around 26-54 days.
Once you have created a broad estimate it is time to do the second round of planning and develop the product backlog items corresponding to epics. The epics can be further broken down into user stories for sprints. Each PBI can be estimated with story points to get a more accurate estimate for the PBIs.
T-shirt estimation is a great way to understand the overall scope of your project. For example, a shopping list that is a small t-shirt size would mean buying a couple of items like a toothbrush and a cola, whereas a large t-shirt size idea would be buying fifty or more items from a shopping list. So, slotting these various tasks into t-shirt sizes will help the team understand the overall scope of the project and what must be accomplished. It helps to understand the effort required by each team member to accomplish the task.
T-shirt sizing, just like Agile, is not a one-size fits all method. Teams must figure out how to use it depending on their project and keeping in mind past projects and retrospectives.
There are some do’s and don’ts for t-shirt sizing that must be followed for success:
Development teams work around this by assigning each size a numerical value such S=1, M=3, L=5, XL=8. Assigning numerical values makes it easier to calculate the velocity. So, if team has completed 2 Small, 4 Medium and 3 Large PBIs then the velocity can be calculated as:
T-shirt estimation allows an extremely fast, almost instant estimation with basic information. Compared to more absolute types of estimation that require more information from stakeholders and users and can result in considerable time consumption and effort, t-shirt sizing is quick and saves close to 80% times in some cases.
T-shirt sizing starts off with the portfolio management team defining the size of the project, and they categorize the project as being extra small, small, medium, large, extra-large etc. The product owner first gets together with the stakeholders and defines a few high-level epics. The epics are given t-shirt sizes based on their perceived complexity. The development team also uses historical data from previous projects to classify tasks into different sized buckets.
T-shirt sizing is a great option for teams new to the whole estimating business. It is fast and simple, and teams can use it till they learn the ropes of the more accurate forms of estimation. Splitting projects into generalized buckets helps the team to break down complex tasks, helps in communication and allows the team to look at a long-term roadmap for the project. When done correctly, t-shirt sizing can boost productivity and save the team a whole lot of effort.
Take Me to the Course
Value Of A Certified Scrum Product Owner
Product Owner vs Scrum Master: Key Differences
Classifications Of The Product Owner Role
18 Aug 2017
01 Feb 2021
25 Apr 2019
09 Apr 2018
22 Jan 2017
20 Dec 2018