Story points measure how much effort a piece of work will take relative to other work - how complicated it is, how much is unknown, and how much of it there is. They are not hours. You compare them to each other, which is what lets a team estimate consistently even though everyone works at a different pace.
Why relative rather than absolute
People are unreliable at estimating duration and considerably better at comparison. Asked how long something takes, answers vary wildly; asked whether it is bigger or smaller than a thing they finished last week, they mostly agree. Story points exploit that: the team calibrates against its own past work, and a velocity emerges from observation rather than from a promise.
What a point is meant to capture
Three things at once, which is why converting to hours loses information.
- Complexity
- How intricate the change is - how many parts of the system it touches and how carefully.
- Uncertainty
- How much is unknown. Two changes of identical size are not the same estimate if one is in unfamiliar code.
- Volume
- How much repetitive work there is. Simple but large still costs time.
The common failure modes
Almost all of them come from treating points as a currency rather than a forecasting aid.
- Converting to hours
- The moment a point equals four hours, it is just hours with an extra step, and the uncertainty it was meant to carry disappears.
- Comparing across teams
- Points are calibrated within a team. One team's five has no relationship to another's.
- Using velocity as a target
- Velocity is a measurement. Made into a goal, it gets inflated, and the estimates stop being useful.
- Estimating an unscoped story
- A story with no acceptance criteria cannot be estimated meaningfully. The number produced will be a guess with a scale attached.
Frequently asked questions
Anything else? Email hello@crowdlinker.com and a person will answer.
Why the Fibonacci sequence?
The widening gaps reflect that precision falls as size rises. Distinguishing a 1 from a 2 is meaningful; distinguishing a 21 from a 22 is not.
Should AI-generated stories come with estimates?
A proposed estimate is a useful starting point, particularly when it is anchored in the actual codebase rather than in the story text alone. It should be reviewable and easy to change - a proposal, not an authority.
How Spectr helps
Write user stories your developers can actually start
Draft stories with real acceptance criteria, then approve, edit or reject each one