Estimation scales
Why Fibonacci cards work in planning poker.
The numbers are not a prediction of duration. They are deliberately uneven choices that make uncertainty easier to discuss.
The point of the gaps
Most planning poker decks use a Fibonacci-like scale: 1, 2, 3, 5, 8, 13, 20, 40, 100. The gap between small values is tight because a team can often distinguish a tiny change from a small one. As work grows, unknowns and possible implementation paths grow too. A wider gap says, “we cannot responsibly claim this is only a little bigger.”
That is why a choice between 5 and 8 is more useful than debating 6 versus 7. It steers the group away from invented precision and toward relative comparison with work it already understands.
Use reference stories, not a conversion chart
Pick a few completed, representative backlog items and remember why they were small, medium, or large. When a new item arrives, ask which reference it resembles and what changes the comparison: unfamiliar technology, a dependency, edge cases, or unclear acceptance criteria. The number belongs to this team’s shared context, so do not turn it into a universal hours-per-point rule.
See what story points mean for a practical way to keep those comparisons team-owned.
When 13 is a signal to reshape the work
A large card is often more valuable as a prompt than as an estimate. It can mean the item needs discovery, a spike, clearer boundaries, or a split into smaller slices that can be inspected in a sprint. There is no sacred cutoff, but agreeing on one helps a team identify work that is too uncertain to carry into a sprint unchanged.