Planning poker vs t-shirt sizing
By Thimira Dulakshitha Kularatne · Updated
Both are ways to size work relative to other work, without guessing hours. The short answer: use t-shirt sizing when you need a quick, rough picture of a big backlog or roadmap, and planning poker with story points when you’re preparing stories for a sprint. Most teams end up using both, at different stages.
How each one works
Planning poker is a way of running estimation: everyone picks a card in private, the cards are revealed at the same time, and the team discusses the differences before agreeing. The cards are usually story points on a Fibonacci-like scale. Read what planning poker is for the full picture.
T-shirt sizing sorts items into buckets: extra small, small, medium, large and extra large. It’s often done quickly out loud or by dragging items into columns, but it works just as well with hidden votes, which avoids everyone copying the first person to speak.
Side by side
| Planning poker (points) | T-shirt sizing | |
|---|---|---|
| Scale | Numbers, usually Fibonacci (1, 2, 3, 5, 8, 13) | Sizes: XS, S, M, L, XL (sometimes XXL) |
| Precision | Finer: good enough to plan a sprint | Coarse: good enough to compare and prioritise |
| Speed | A minute or two per story | Seconds per item; dozens in one session |
| Adds up? | Yes. Points sum into velocity | No. You can’t add an M to an L |
| Best for | Refined user stories before a sprint | Epics, roadmaps, early backlog grooming |
| Main risk | False precision, points treated as hours | Too vague to plan capacity |
When to use which
- Roadmap and quarterly planning: t-shirt sizes. You’re comparing epics and ideas that haven’t been broken down, so precise numbers would be made up anyway.
- Backlog grooming with lots of items: t-shirt sizes to triage quickly, then points for the ones that make it into the next few sprints.
- Sprint planning: planning poker with points. You need numbers that add up so you can compare the plan against the team’s velocity.
- Teams new to estimation: start with t-shirt sizes to build the habit of relative sizing, then switch to points.
Mapping t-shirt sizes to story points
If you size epics in t-shirts and stories in points, an agreed mapping keeps the two consistent. This is a common starting point; adjust it to your team:
| Size | Points | Roughly means |
|---|---|---|
| XS | 1 | Trivial, well understood |
| S | 2–3 | Small, one area of the code |
| M | 5 | A typical story with a few moving parts |
| L | 8 | Big; consider splitting |
| XL | 13 | Too big for one sprint: split it |
Treat the mapping as a translation for planning, not a rule. When an epic is broken into stories, estimate those stories fresh rather than dividing the epic’s size between them. More on the numbers in Fibonacci story points explained.
Questions
Is t-shirt sizing a form of planning poker?
It can be. Planning poker is the process: private votes, a simultaneous reveal, then discussion. T-shirt sizes are just a different deck. Many teams play planning poker with t-shirt cards for rough sizing.
Can you calculate velocity with t-shirt sizes?
Not directly, because sizes aren’t numbers. If you need velocity, either convert sizes to points with an agreed mapping or re-estimate the stories with a numeric deck once they’re refined.
Which is better for a new team?
T-shirt sizes are easier to start with, because nobody argues over numbers. Once the team is comfortable comparing work, move sprint-level stories to Fibonacci points so you can track velocity.
Try both in one room
Benchmark has a T-shirt deck (XS to XXL) as well as Fibonacci, modified Fibonacci and powers of 2. Switch decks from the room settings between agenda items, and everyone’s cards stay hidden until the reveal either way.