Sprint velocity tracking

Sprint velocity tracking is the practice of measuring how much work a team completes each sprint and following that figure over time to forecast future delivery. A single sprint result is noisy, so the value comes from the trend. If a team completes 32, 28 and 30 story points across three sprints, its average velocity is 30, and that average is what you plan the next sprint against.

Velocity is a capacity signal, not a productivity score. It tells you roughly how much committed work this specific team can finish in a sprint, given its current size, skills and interruptions. It is meaningless to compare one team velocity against another because the points scale is local to each team.

Velocity for a single sprint is the sum of the estimates of all items completed and accepted in that sprint. To make it useful for planning, take a rolling average over the last few sprints, typically three to six, so one unusual sprint does not distort the figure.

For example, with completed totals of 24, 30, 27 and 31 points over four sprints, the four-sprint average velocity is 28. You then plan the next sprint around 28 points, adjusting down for planned leave or up if the team is growing.

1. 1

   Define what counts as done

   Agree a clear definition of done so only finished, accepted work is counted in a sprint total.

2. 2

   Sum completed work per sprint

   Add up the estimates of every item that met the definition of done within the sprint.

3. 3

   Average over recent sprints

   Take the mean of the last three to six sprint totals to smooth out single-sprint noise.

4. 4

   Plan against the average, not the peak

   Use the rolling average for forecasting, and adjust for known capacity changes such as leave or new joiners.

Velocity rising or falling is a symptom, and a metric tree exposes the cause. Decompose velocity into the drivers a team can act on: available capacity, how much of that capacity is interrupted, the quality of estimates, and the flow of work through the sprint. A dip is rarely about effort. It is usually fewer available days, more unplanned work, or estimates drifting.

KPI Tree lets you connect each driver to the team and the actions that influence it, with RACI ownership on every branch. When velocity moves outside its normal range, the change can be pushed to the accountable owner alongside the specific driver responsible, so the conversation is about interruptions or capacity rather than a vague sense that the team slowed down.

Absolute velocity numbers cannot be benchmarked across teams because point scales differ. What can be benchmarked is the stability of the trend. A predictable, low-variance velocity is the goal, since it is what makes forecasting reliable.

Improving velocity tracking is about stability and honesty, not bigger numbers. The aim is a figure you can plan against, so reduce the things that make it swing and resist the temptation to inflate estimates. A steady velocity beats a high one that you cannot trust.

1. 1

   Treating velocity as a target

   When velocity becomes a goal, teams inflate estimates and the number stops reflecting real capacity.

2. 2

   Comparing velocity across teams

   Point scales are local to each team, so cross-team velocity comparisons are meaningless and damaging.

3. 3

   Counting unfinished work

   Including partially done items in the total makes velocity look healthy while delivery quietly slips.

4. 4

   Planning against a single sprint

   Using one good or bad sprint instead of a rolling average leads to over-commitment or sandbagging.