StackLift
All terms

Glossary

Story points

A unit of relative effort used to size software work instead of raw hours.

Story points are a relative measure of the effort, complexity, and uncertainty in a piece of software work. Instead of guessing exact hours, teams compare a task to known reference work and assign points, which produces more honest, comparable estimates over time.

Points capture three things at once: how much work it is, how complex it is, and how risky or unknown it is. A simple form might be 1-2 points; a payments integration with edge cases might be 13.

StackLift prices delivery in story points so cost maps to deliverable complexity, not to time spent - which keeps estimates predictable and removes the incentive to pad hours.

Start in 60 seconds

Put this to work on your project.

Get an AI-assisted, architect-reviewed scope, cost range, and timeline. Free, no login.

The complete guide

Story points: the complete guide

Updated November 18, 2026 ยท Published by , Founder, StackLift AI. Written with AI assistance; how we make content.

Short answer

Story points are a relative unit for sizing software work. Instead of guessing hours, a team scores each piece of work by comparing its effort, complexity and uncertainty to work it already understands. The scores add up to a total that supports planning, pricing and progress tracking without pretending to know exact durations.

What are story points in simple terms?

Think of T-shirt sizes. You may not know how many minutes a task takes, but you can say it is smaller than a medium one. Story points turn that comparison into numbers, so the sizes can be added up and tracked.

A point is not an hour, a day or a dollar. It measures how big a piece of work is relative to other pieces of work. Two teams can give the same feature different scores, and that is fine, because each team scores against its own reference work.

The word story comes from the user story, a short description of something a user needs, such as "a customer can reset a password." Points size that story. For how sizing fits into a whole project plan, see software estimate.

How do story points work?

Teams pick a scale that grows in steps, often a Fibonacci-style sequence. The widening gaps are deliberate. They stop people from arguing over whether something is a six or a seven when the honest answer is that it is large and uncertain.

  1. Pick a reference item the team knows well and call it a small score.
  2. Compare each new story to the reference: is it about the same, twice as big, or much bigger?
  3. Each person scores privately, often with planning poker, so one loud voice does not anchor the group.
  4. Talk through the biggest gaps. The person who scored high usually knows about a risk the others missed.
  5. Settle on one score, and split anything too big to finish in a single sprint.

Over time the team learns how many points it finishes in a sprint. That figure is called velocity. It helps forecast how long a backlog will take, as a range rather than a promise.

What do story points actually measure?

A score blends three things. The first is the amount of work. The second is complexity, meaning how many moving parts, rules and integrations are involved. The third is uncertainty, meaning how much is unknown and could go wrong.

Two stories that take similar typing time can score very differently
StoryAmount of workComplexityUncertaintyLikely score
Add a field to an existing formSmallLowLowSmall
Build a login screen with a standard identity providerMediumMediumLowMedium
Sync orders with a partner API you have not used beforeMediumHighHighLarge
Payment flow with refunds, retries and tax rulesLargeHighHighVery large, usually split

Story points vs. hours: what is the difference?

Hours try to predict how long one person will spend. Points describe how big the work is, whoever does it. Hour estimates tend to look precise and still miss, because they hide uncertainty inside a single number.

Story points compared with hourly estimates
Story pointsHours
What it describesRelative size of the workPredicted time for one person
Handles uncertaintyYes, it raises the scoreUsually hidden in a padded guess
Depends on who does the workMuch lessHeavily
Cost model it supportsPriced by deliverable complexityTime and materials billing
Main riskTreated as a hidden time measureRewards slow work and padding

If you are weighing the billing models, the fixed scope vs. hourly billing comparison walks through the trade-offs.

Why do story points matter when you buy or build software?

When a vendor prices by hours, you pay for time spent, so a slower team earns more. When pricing follows deliverable complexity, the price tracks what you receive. That aligns the vendor with finishing the work well.

Points also make change visible. If a new request adds points to the baseline, everyone can see how much it adds before work starts. A request that adds nothing is easy to approve, and one that adds a lot starts a conversation. This is what a change request records.

Points also show you where risk sits. A large score on a small feature tells you the team sees an unknown, such as an unfamiliar integration, and you can ask about it before you sign. The software estimation playbook shows how to read an estimate this way.

What are the common mistakes with story points?

  • Converting points to hours with a fixed rate. That turns them back into a time estimate and removes the benefit.
  • Comparing velocity between teams. Each team has its own scale, so the numbers do not line up.
  • Using points to rate people. Once a score becomes a target, teams inflate it.
  • Scoring work nobody has understood yet. Unclear stories get padded scores, which hides the real problem.
  • Changing the scale mid-project. The history no longer predicts anything.
  • Skipping the conversation. The discussion about the gap between two scores is where most of the value is.

Points are also not a promise. They support a forecast, and a forecast improves as the team learns the codebase and the client clarifies what they need.

How does StackLift use story points?

StackLift prices by deliverable complexity in story points, not by hours. Scope, budget and delivery path are agreed before engineering begins, and every estimate is reviewed by a senior architect before it becomes a commitment.

The free project estimate scopes a project feature by feature and returns cost, timeline and team in about five minutes, with no signup. If you want a deeper plan first, the AI Scope Sprint produces detailed scope, user stories, tech stack, and a cost and timeline range in three days.

Because the estimate, proposal, project and client portal share one baseline, any later change is explicit and shown against the points you agreed to.

Which scale should you use for story points?

Teams need a scale that gets less precise as items get bigger, because large work is harder to judge. The Fibonacci-style sequence (1, 2, 3, 5, 8, 13) is popular for that reason: the gaps grow, so nobody argues over whether something is a 14 or a 15.

Common sizing scales
ScaleHow it looksGood for
Fibonacci-style1, 2, 3, 5, 8, 13Sprint-level stories where you want to track pace
Powers of two1, 2, 4, 8, 16Teams who want fewer, wider buckets
T-shirt sizesXS, S, M, L, XLEarly, rough sizing of epics or a whole roadmap
Plain countEach story counts as oneTeams whose stories are all kept very small

Whatever you pick, define it once and keep it. A scale only works as a shared language if everyone reads a 5 the same way, and if you resist the urge to convert it quietly back into days.

How do teams estimate story points together?

The most common technique is planning poker. Each person privately picks a card, everyone reveals at once, and the team discusses the gaps. The value is less in the number than in the conversation, because a spread between a 2 and a 13 usually means two people imagine different work.

  1. Pick a reference story the team knows well and give it a value, such as 3.
  2. Read the next story aloud and let people ask questions about scope and risk.
  3. Everyone selects a card without discussing it first, so early opinions do not sway others.
  4. Reveal together. If the numbers agree, record it.
  5. If they differ, the highest and lowest voters explain their reasoning, then the team votes again.
  6. If the story is still too large or too vague, split it or add questions to the backlog instead of guessing.

Quicker alternatives, such as affinity sizing, where the team sorts a large pile of stories into columns by relative size, work well when you have dozens of items to estimate at once.

What is velocity, and how does it forecast delivery?

Velocity is the number of story points a team finishes in a sprint. After a few sprints you see a typical range, and you can divide the remaining points in the backlog by that range to forecast how many sprints are left. The forecast is a range, not a promise, and it improves as more sprints complete.

Using velocity responsibly
DoDo not
Use it to plan one team's next sprintCompare it between teams; each scale is its own
Look at a trend over several sprintsReact to one unusually high or low sprint
Count only work that meets the definition of doneCount partly finished stories
Re-forecast when scope changesHold the old date after the backlog grows

Velocity changes when the team changes, when the product gets harder to work in, or when a lot of technical debt builds up. A steady drop is a signal worth discussing at the retrospective.

Story points compared with other estimation methods

Estimation methods side by side
MethodWhat it estimatesStrengthWeakness
Story pointsRelative size and complexityResistant to individual speed and to false precisionTakes a few sprints to become useful
Hours or daysEffort for one person or teamEasy to understand and tie to a calendarLooks exact but is usually optimistic
T-shirt sizesRough relative sizeFast for a first roadmapToo coarse for sprint planning
No estimatesCounts stories flowing throughVery little ceremonyNeeds consistently small stories

No method removes uncertainty. The goal is to make it visible early, which is why the software estimate pairs sizing with assumptions and risks. The software estimation playbook goes deeper on that.

What do small and large stories look like?

Examples help calibrate a scale. These are illustrative sizes; a real team sets its own reference stories.

  • Small: change the label and validation message on an existing form field.
  • Small to medium: add a sortable column to an existing table, with a test.
  • Medium: add password reset by email, including the email template and expiry rules.
  • Large: accept card payments through a payment provider, with refunds and failure handling.
  • Too large, so split it: build the reporting module. Split by report, by role or by data source.

Notice that points reflect not only code but also uncertainty and testing. Work touching an unfamiliar third-party system earns a higher number than familiar work of the same visible size.

How do you know your story point estimates are any good?

  • Stability: finished points per sprint stay within a predictable band.
  • Fewer surprises: stories rarely turn out far bigger than first sized.
  • Small stories: most items fit comfortably inside one sprint.
  • Shared understanding: planning discussions end with agreement, not stalemates.
  • Honest re-estimation: scope changes are re-sized, not absorbed silently.

If these are failing, look at the stories before the scale. Unclear acceptance criteria and oversized stories cause far more bad estimates than the choice of numbers.

What tools support story points?

Most backlog tools have a story points field and velocity reports, including Jira, Linear, Azure DevOps, GitHub Projects and ClickUp. Free planning poker apps and plugins exist for distributed teams. Choose one that keeps estimates beside the story and its acceptance criteria so the reasoning is not lost.

Questions people ask

How many story points is a day of work?

There is no fixed conversion, and that is the point. A point measures relative size, not time. Teams that force a points-to-hours rate end up with hour estimates under a different name.

What scale should I use for story points?

A Fibonacci-style sequence is the most common because the gaps grow as items get larger. Any scale works if the team uses it consistently and splits items that are too large to finish in one sprint.

Can non-developers understand story points?

Yes. Treat them as a size rating, like small, medium and large with numbers on top. You do not need to know how the code works to see that a large rating means more work or more unknowns.

What is velocity?

Velocity is the number of points a team completes in a sprint. Averaged over several sprints, it helps forecast how long the remaining backlog will take. It is a planning aid, not a performance score.

Are story points the same as function points?

No. Function points are a formal method that counts inputs, outputs and data in a specified way. Story points are a team's own relative scale, agreed through discussion.

Why would a vendor price by story points instead of hours?

Pricing by deliverable complexity ties what you pay to what you receive, and it removes the reason to pad time. You agree scope and budget up front, and changes show as additions to the baseline.

Is a story point equal to an hour?

No. A point is a relative measure of size and complexity, not time. Converting points to hours defeats the purpose and brings back the false precision they were meant to avoid.

Who should estimate story points?

The people who will do the work: developers, testers and designers on the team. Estimates handed down by managers or sales tend to be optimistic and are not owned by the people building.

Can you use story points on a fixed-price project?

Yes. Points measure the complexity of the scope, and that scope can be priced. StackLift prices by deliverable complexity in story points, with scope, budget and delivery path agreed before engineering begins.

Sources

Sources and further reading

The standards, rules, research and official documentation this page draws on.