Understanding Sprint Velocity
If you've ever been asked "When will it be done?" and struggled to give a confident answer, you're not alone. Sprint velocity is one of the most powerful tools you have to answer it with data instead of guesswork.
What Is Velocity, Really?
Velocity is the amount of work a team completes in a sprint, measured in story points.
That's it. No magic, no complexity. It's a trailing indicator that tells you how much your team has historically delivered, which becomes the foundation for predicting future performance.
Think of velocity like your average driving speed. If you typically drive 60 miles per hour on the highway, you can reasonably estimate that a 300-mile trip will take about 5 hours. You wouldn't promise to do it in 3 hours just because you feel optimistic, nor would you beat yourself up if traffic slows you down occasionally. Velocity works the same way—it's a baseline based on reality, not ambition.
Why Velocity Matters
Predictability Over Speed
Stakeholders don't just want fast delivery—they want predictable delivery. A team that consistently delivers 40 story points per sprint is more valuable to plan around than a team that delivers 60 one sprint, 20 the next, and 45 the one after.
When you know your velocity, you can:
- Forecast release dates with confidence intervals ("We'll likely ship in 6-8 sprints")
- Push back on unrealistic deadlines with data ("Our velocity shows this is a 12-sprint commitment, not 6")
- Identify problems early when velocity drops unexpectedly
Building Stakeholder Trust
Nothing erodes trust faster than missed commitments. When you consistently under-promise and over-deliver based on your actual velocity, stakeholders learn to trust your estimates. That trust translates into less micromanagement, fewer emergency meetings, and more autonomy for your team.
Smarter Planning
Velocity helps you avoid the planning fallacy—the cognitive bias where we underestimate how long things will take. By grounding sprint planning in historical performance instead of optimism, you set your team up for sustainable success.
How to Calculate Velocity
The Basic Formula
Velocity = Sum of story points completed in a sprint
A "completed" story meets your team's definition of done—coded, tested, reviewed, and potentially shippable. Partially done work counts as zero.
Calculating Average Velocity
Most teams use a rolling average of the last 3-6 sprints:
Average Velocity = (Sum of story points from last N sprints) / N
Example Calculation
Here's a real velocity calculation for a team over six sprints:
| Sprint | Story Points Completed | Notes |
|---|---|---|
| Sprint 1 | 38 | New team member joined |
| Sprint 2 | 42 | Standard sprint |
| Sprint 3 | 35 | Two developers out sick |
| Sprint 4 | 44 | Strong sprint, low interruptions |
| Sprint 5 | 40 | Standard sprint |
| Sprint 6 | 39 | Standard sprint |
| Average | 39.7 | Round to 40 for planning |
For Sprint 7 planning, this team should commit to approximately 40 story points. Not 50 because they "want to do better." Not 30 because they're being cautious. Forty, because that's what the data says.
Using Velocity for Release Planning
If your backlog has 240 story points remaining and your velocity is 40:
Sprints needed = 240 / 40 = 6 sprints
With 2-week sprints, that's approximately 12 weeks (3 months) to completion.
Pro tip: Always add a buffer. Use velocity ranges rather than single numbers. If your velocity ranges from 35-45, communicate that the release will take 5.5-7 sprints (11-14 weeks).
Common Velocity Mistakes
1. Comparing Teams
Never compare velocity across teams. Team A's "8 points" is not the same as Team B's "8 points." Story points are relative estimates meaningful only within a single team.
2. Ignoring Outliers
That sprint where you only completed 15 points because half the team had the flu? Or the 70-point sprint where you cleaned up a year of tech debt? These outliers can distort your average. Consider excluding them or using median instead of mean if your velocity varies wildly.
3. Not Adjusting for Holidays and PTO
If your average velocity is 40 with a full team, don't expect 40 during holiday weeks. Adjust expectations proportionally:
Adjusted Velocity = Normal Velocity × (Available Team Capacity / Full Capacity)
With 3 of 8 developers out: 40 × (5/8) = 25 story points
4. Treating Velocity as a Target
Velocity is a measurement, not a goal. When teams are pressured to "improve velocity," they game the system—inflating estimates, cutting corners, or carrying stories over as "almost done."
5. Including Incomplete Work
A story that's 90% done is 0% done in velocity terms. Until it's fully complete per your definition of done, it doesn't count. No partial credit. This rule seems harsh, but it keeps the metric honest.
Improving Velocity (The Right Way)
If velocity isn't about working harder, how do you improve it?
Remove Blockers
The fastest way to increase velocity is to eliminate what slows your team down:
- Long code reviews? Implement pairing or smaller PRs
- Frequent production fires? Invest in monitoring and stability
- Unclear requirements? Spend more time in refinement
- External dependencies? Identify them early and escalate blockers
Reduce Context Switching
Nothing kills velocity like constant interruptions. Protect your team's focus time:
- Batch meetings into specific days
- Use async communication where possible
- Shield the team from non-urgent requests during sprints
Address Technical Debt
That "quick fix" from six months ago is now slowing down every feature that touches that code. Dedicated time for refactoring pays velocity dividends over time.
Right-Size Your Stories
Stories that are too large (13+ points) create uncertainty and often roll over between sprints. Stories that are too small (1 point) create overhead. Aim for a healthy mix with most stories in the 3-8 point range.
Velocity in SprintFlint
Tracking velocity manually in spreadsheets works for small teams, but it quickly becomes tedious and error-prone. SprintFlint's velocity dashboard does the heavy lifting:
- Automatic calculation from completed work
- Trend visualisation over time
- Outlier detection and anomaly highlighting
- Capacity-adjusted forecasts based on team availability
Start Your First Sprint →
The Bottom Line
Velocity isn't a score to maximize—it's a lens to understand your team's capacity. Used correctly, it transforms estimation from art to science, builds stakeholder trust through consistent delivery, and helps you identify systemic problems before they become crises.
Remember: Start measuring it honestly. Calculate it consistently. And most importantly, use it to have better conversations about what's realistic, not to push your team harder.
The best engineering teams aren't the ones with the highest velocity. They're the ones whose velocity you can count on.