Fibonacci vs T-Shirt Sizing: Which Estimation Scale Should You Use?

Every planning poker session needs a scale. The two most popular are the modified Fibonacci sequence (1, 2, 3, 5, 8, 13, 21) and T-shirt sizes (XS, S, M, L, XL, XXL). Both are relative scales, but they suit different situations.

Why not just use 1 to 10?

A linear scale suggests a precision nobody has. Is a task a 6 or a 7? Nobody can tell, and the team wastes time debating it. Scales with growing gaps force a more honest question: is this roughly the same as that, or clearly bigger? That's also why the jumps get wider as numbers grow. Uncertainty grows with size.

The Fibonacci scale

Each number is roughly the sum of the two before it. Most teams use a "modified" version that rounds large values (20, 40, 100) and adds a "?" card.

Best for

• Sprint-level estimation of user stories
• Teams that track velocity and use it for forecasting
• Mature backlogs where stories are already well understood

Watch out for

• Numbers that quietly turn into hours
• Comparing velocity between teams. Points are only meaningful within one team

T-shirt sizes

T-shirt sizes trade numeric precision for speed and approachability. Non-technical stakeholders understand "this is an L" immediately.

Best for

• Roadmap and epic-level sizing
• Early discovery, when details are fuzzy
• Workshops that include people outside the dev team

Watch out for

• Sizes can't be added up for velocity without mapping them to numbers
• Teams sometimes disagree on what "M" means; agree on reference items

Quick comparison

Our recommendation

Use T-shirt sizes while shaping epics, then switch to Fibonacci story points once stories are ready for a sprint. Planno supports both, plus a simple 1–10 deck, and the host can switch decks in one click. Try it free.