# Team Velocity

Team velocity is the amount of estimated or counted work one team completes within an iteration, interpreted from that team's recent history.

- Type: metric
- Domain: Delivery & agility
- Level: Foundational
- Last reviewed: 2026-09-22
- Original: [Soroush’s Product Wiki](<https://wiki.soroush.uk/topic/team-velocity/>)

## What it means

Team velocity records how much work a particular team finishes in an iteration. It commonly sums story points for items meeting the team's done criteria, though some tools can count items or other estimates. Because points and practices are locally defined, velocity is meaningful primarily within the same stable team and context.

## Why it matters

A recent velocity range can help a team choose a plausible amount of work for an upcoming sprint and notice changes in its planning system. It can also support rough release projections when scope and team conditions are sufficiently stable. Its purpose is forecasting and learning, not proving productivity or business impact.

## When to use it

Use velocity with iterative teams that estimate consistently enough for their own history to remain interpretable. Review several recent sprints and preserve variation instead of relying on one number. Recalibrate after membership, workflow, definition-of-done, estimation, or product-context changes that make earlier observations less representative.

## How it works

1. Agree which items and estimation unit count, and count work only when it satisfies the team's done criteria.
2. Record completed work for each sprint, assigning spillover to the sprint in which it actually finishes.
3. Inspect a recent range or distribution rather than treating the arithmetic average as guaranteed capacity.
4. Use the history as one input to sprint planning alongside availability, goals, risks, and dependencies.
5. Discuss changes with the team and pair velocity with outcome, quality, and flow measures.

## A practical example

A team completed 24, 31, 27, 18, and 29 points in its last five sprints. Rather than declaring velocity to be 26 forever, it considers the range, notes that the 18-point sprint included an outage, and selects work around a coherent sprint goal. It does not compare its points with another team.

## Where it breaks

Velocity inherits uncertainty from subjective estimates and can change when teams learn, split work differently, or alter membership. It does not measure value, quality, efficiency, or individual contribution. Targets create pressure to inflate estimates or manipulate scope. Averages also conceal variation and should not become commitments to exact output.

## Common mistakes

- Comparing velocity across teams whose point scales, work, and definitions of done differ.
- Setting velocity as a performance target, which invites point inflation and gaming.
- Counting partially finished work or assigning spillover to the sprint in which it was planned.

## Sources

1. [View and Configure Team Velocity](<https://learn.microsoft.com/en-us/azure/devops/report/dashboards/team-velocity?view=azure-devops>) — Microsoft Learn · Technical documentation
2. [Sprint Velocity in Scrum](<https://www.atlassian.com/agile/project-management/velocity-scrum>) — Atlassian · Professional guide

## Related topics

- [Throughput](<https://wiki.soroush.uk/topic/throughput/>)
- [Burnup Chart](<https://wiki.soroush.uk/topic/burnup-chart/>)
- [Burndown Chart](<https://wiki.soroush.uk/topic/burndown-chart/>)
