Guide · Agile estimation
Story points: how to calculate and point stories
What a story point means, how agile teams estimate them, how they work in Jira, and how to turn points into days when someone asks for a date.
In short. To calculate story points, pick a reference story the team knows and call it a 3. Then size each new story against it on the Fibonacci scale (1, 2, 3, 5, 8, 13, 21). Vote at the same time with planning poker, and talk about any big spread before you settle.
What are story points?
A story point is a relative unit of effort. A 2-point story is roughly twice the work of a 1-point story, and a 5 is a bit more than twice a 2. Points leave hours out on purpose, so the team compares stories with each other instead of guessing time.
Points cover everything that makes a story big: the amount of work, how complex it is, and how much nobody knows yet.
Why story points use the Fibonacci scale
Most teams pick from 1, 2, 3, 5, 8, 13 and 21. The gaps grow with the numbers because bigger work is harder to size precisely: nobody can tell a 13 from a 14, but everyone can tell a 3 from a 13.
If a story keeps landing on 13 or above, it usually needs splitting before anyone commits to it.
How to calculate story points
- Pick a reference story the whole team remembers, and call it a 3.
- Compare each new story against it. Bigger or smaller than the reference? By how much?
- Vote at the same time with planning poker, so nobody anchors on the first number said out loud.
- Talk about the spread, then vote again. When the votes are far apart, someone knows about work the others missed.
- Record the points where the team tracks its work. In Jira team-managed projects, that’s the Story point estimate field.
How to point stories as a team: a worked example
Here’s how those steps look on a short backlog. The team and its stories are made up. Its reference story, the 3, is “Send a password reset email”: one form, one email, one link that expires. Everyone remembers building it, so every new story gets compared with it.
Write the stories the same way before you vote, so everyone is sizing the same thing. The user story template has the format, the acceptance criteria and the questions to settle first.
| Points | Story and reasoning |
|---|---|
| 1 | Change the wording of the invite emailMuch smaller than the reference: a text change, with no new logic to test. |
| 2 | List pending invites on the members pageSmaller: it shows data the app already has and sends nothing. |
| 3 | Resend an inviteThe same size as the reference: one button, one email, one link that expires. |
| 5 | Invite a teammate by emailBigger: a form, an email and an accept step, plus the case where the person already has an account. |
| 8 | Invite a whole team from a CSV fileMuch bigger and the least known: bad rows, duplicates, and what happens when only some invites go through. |
The 8 is the story to talk about. If one person votes 5 and another 13, ask each of them what they’re counting. Had it settled on 13, it would usually need splitting before anyone commits to it. Run the rounds with planning poker online, and nobody hears a number before picking their own.
What a story point includes
A point covers three things at once, so a story can be big for more than one reason:
- Effort: how much work there is to do.
- Complexity: how hard that work is to get right.
- Uncertainty: how much nobody knows yet.
A long, repetitive change is mostly effort. A small change to code nobody understands is mostly uncertainty. Both can land on the same number.
Points are also relative. They mean something only next to a team’s own reference stories and pace, so a 5 on one team is not a 5 on another, and points don’t compare across teams.
Points are one way to size work. Software estimation methods covers the others, and when each one fits.
Story points in Jira
Jira has a field for story points on each issue; in team-managed projects it’s called Story point estimate. Fill it in during planning, and the velocity chart shows how many points the team finishes each sprint.
That velocity is what lets you turn points into time, which is what the calculator below does.
Planning the next sprint? The sprint planning template puts the goal, capacity and backlog on one page.
Calculator
Story points to days.
Enter your team’s numbers to see what a point means in days for your team, on average.
We divide the person-days in a sprint by the points you usually finish.
- 1 point1.3person-days
- 2 points2.7person-days
- 3 points4person-days
- 5 points6.7person-days
- 8 points10.7person-days
- 13 points17.3person-days
If all 4 people worked on it together, a 13-point story would take about 4.3 working days.
This is an average. Your 5-point stories don’t all take the same time: some finish in half of it, some take twice as long. That spread is what a single number hides, and it’s why dates built on velocity slip.
When days work better than points
Points are good for comparing stories and tracking how much a team gets through. They’re weak when someone needs a date, because velocity is an average and averages hide how much any single item can vary.
For dates, estimate each part of the work as a best, likely and worst case in days, and choose how sure you need to be. Many teams keep points on the board and use ranges for commitments.
| Aspect | Story points | Day ranges |
|---|---|---|
| Measures | Size relative to other stories | Dev-days, best to worst case |
| Best for | Sizing the backlog, tracking velocity | Dates, commitments, finding the risky parts |
| “When will it ship?” | Only through average velocity | A date with a confidence level, like 80% |
FAQ
Story points FAQ
How do you calculate story points?
How do you convert story points to hours or days?
Why do story points use Fibonacci numbers?
Can you compare story points between teams?
Where do story points go in Jira?
Need a date, not points?
SprintProof breaks a feature into parts, estimates each as a range in days, and tells you the date you’re 80% sure of. It works alongside the points already on your Jira board.