Story points are a way for a team to size work by how hard it is, not how long it takes. Get the idea right and sprint planning becomes a short conversation; get it wrong and the points turn into a timesheet nobody trusts.
What a story point measures
A story point is a relative unit of effort for one piece of work, usually a user story in the backlog. It rolls three things into a single number:
- Effort: how much work there is. Ten near-identical form fields take longer than one.
- Complexity: how hard the work is to get right. Tricky logic, awkward edge cases, lots of moving parts.
- Uncertainty: how much you don't know yet. An unfamiliar API, legacy code nobody has touched in years, a requirement still being argued over.
A story point is not an hour, and not a day. It has no fixed size on its own. It only means something next to other stories the same team has sized: a 4-point story should feel like about twice the work of a 2.
Why not estimate in hours
Hours sound more precise, but they cause three problems.
First, people are poor at absolute estimates and fairly good at comparisons. 'How long will this take?' invites a guess. 'Is this bigger or smaller than the login page we built last month?' is a question most developers can answer quickly and well.
Second, hours depend on who does the work. A senior developer who knows the billing code might need a day; someone new to it might need three. The story is the same size either way, and points let the team agree on that size without first deciding who picks it up.
Third, an estimate in hours tends to become a deadline. Once 'six hours' is written on a ticket, a story that takes ten looks like a failure, even when the extra four went on a bug nobody could have seen coming. Points keep the estimate about the work and leave forecasting to velocity, covered below.
The scale: why 1, 2, 3, 5, 8, 13
Most teams don't use every number. They pick from a sequence based on Fibonacci, where each value after the first two is the sum of the two before it: 1, 2, 3, 5, 8, 13, and in many planning poker decks 20, 40 and 100 after that.
Wondering how this was made? How we make these