Agile Notes (No. 52)
Hello all! Welcome new subscribers! Thanks for reading. We continue with Collaboration Explained by Jean Tabaka. I honestly was just not feeling it this week. Nothing really jumped out at me so we have skipped ahead to Chapter 14: Team Estimating Approaches.
Don’t get me wrong - there is a bunch of good, structural stuff in the chapters from where we left off in Chapter 9. All required a lot of lengthy explanations and I felt would bog us down.
So….this week….T-Shirt Sizing. I have never used this with a team for estimating purposes. I have had project/program estimates forced upon me and my teams from upper management. Spoiler alert: they were NEVER accurate. (This happens when you have non-technical people making WILD assumptions.)
I like the guidance Tabaka provides about starting with your best guess at what a medium-sized project looks like and then defining a Small project as half the size of the Medium and a Large is double the size of the Medium. It is a solid starting point.
I also love the note she provides! No XL, XXL and on! You MUST break those big ones into smaller efforts. I did this with some of my Scrum teams when estimating Story Points. I coach nothing over a “5”. Break it up into smaller chunks. The nature of complex technical environments combined with complex code ALWAYS resulted in unanticipated challenges. That “5” often easily bumped to a “10”.
I wish Jira had the ability to T-shirt size items in the Backlog. Or does it and I just don’t know about it? Or are there other applications out there that will do this? Let me know in the comments.
Thanks for reading! Please like, share, subscribe and comment!
I work in the area of Lean Portfolio Management ( We work with SAFe). We use T- Shirt sizing extensively at the portfolio to get a feel for the medium term investment and capacity allocations. Your point about no XXL is spot on. This is the equivalent of saying I’ll take the whole ART for 3 PIs please. Not helpful.
My guidance at the portfolio level is:
1. Ensure your Epics are clearly linked to a short to medium term (1-3 PIs) objectives / KRs.
2. Make them small enough to give meaningful insight to the plan, ie. Have enough granularity that you can plot some outcomes that are linked to objectives.
3. Use your Solution Architects in planning and setting T-Shirt sizes. As a dev would with Story points - in other words, leverage their experience.
Recently we have been running experiments to try and calibrate the Epic T-shirt with feature volumes and velocity. Early days.
On Jira, not sure about out of the box, but it didn’t take much for our admin to configure our Portfolio Epic issue types with T-Shirt sizing fields.