Use relative sizing to estimate stories in an agile project. T-shirt sizes provide a way to categorize and estimate buckets of stories using sampling.

Page content

T-shirt sizing is relative and less prone to errors than estimation of effort, in real days. For instance, you can be guaranteed that two developers are like two watches. They never agree when it comes to how long it will take to develop a piece of functionality. One might say that it will take two weeks while the other smirks with an I-can-do-it-in-two-days attitude. That may be true as different developers come with varied skills and experience.

Relative sizing is about comparing two different pieces of functionality and agreeing on which one is smaller (or larger) in size and by determining how much smaller. Rarely do developers disagree on this. And if they do, you can be sure that the story is not clear enough or there are major unknowns. One way to estimate using relative sizing is to use the T-shirt sizes; small, medium, large and x-large. Too many sizes may confuse the developers and the team, so it is wise to stick to three or four sizes. You always estimate with more than one developer in the room, so that there can be a good debate about confusing points. The skills and experience levels of the developers are not that important. It may even be good to have a wide spectrum.

Start by writing up cards with S, M, L, XL for each of the developers in the room. The BA then explains the stories, one by one, and the developers estimate by raising the size-card together so that one does not influence the other. Whenever there is a mismatch in their estimates they discuss and debate the point in the room and re-estimate right away. Usually it ends up with a consensus, as debates clear up the confusion. Always estimate with the end-users and stakeholders in the room to clear up any doubts.

At the end of this exercise all the stories would have been classified into the different size buckets. Now what? Sample and figure out the time required to develop a small, medium, large and x-large. Start with one from each size-bucket and break them down into developer tasks, at the minute level possible. It is not too much to break down to a task level of even an hour’s work. Then estimate each of these minute tasks will calculate how long it will take, in real hours, to develop a small, a medium and so on.

Sampling is quite important in this process. It is good to start with random samples and then move on to boundary samples. From each bucket, pick ones that seem to be the largest and the smallest. Estimate these. How much do they vary? Is the smallest medium larger than the largest small? By how much? What is the variance within each bucket? Can the smallest and the largest in mediums still be called mediums?

Depending on the results of the above analysis, the buckets and the stories in them may have to be rearranged.

Once this exercise is over, we can now apply the real estimates from the samples to all the stories in a bucket to get an overall estimate of the size of the project. It is now time to step back and take stock of the budget, cost-benefit, approvals, schedule, road map and so on before progressing any further.

This post is part of the series: Project Estimation in an Agile World

The estimation process in an agile project before you plan; how to capture stories, estimate with very little information, yet be accurate. Applying science to capture good estimates.

  1. Estimating the Development Effort for your Software Project
  2. T-Shirt Sizing to Estimate Development Effort in a Software Project
  3. Risk Adjusted M-L-M Estimation for Software Projects