Board completion velocity

Board completion velocity is the rate at which items on a board are moved to done over a fixed period of time. If a board completes 30 cards in a two-week period, its velocity is 15 cards per week. The point of the metric is not the raw count but the trend: a velocity that drifts down week after week tells you the board is losing momentum even if it still looks busy.

Velocity differs from throughput in framing. Throughput is the count of finished work, and velocity expresses that count as a rate so you can compare periods of different lengths and forecast how long the remaining backlog will take. A board with 60 open cards and a velocity of 15 per week has roughly four weeks of work left, assuming nothing new arrives.

Pick a consistent period, then count every item that reached the done state within it. Divide by the length of the period to get a rate. If a board completed 45 cards over three weeks, the velocity is 15 cards per week.

For forecasting, take an average velocity across several recent periods rather than a single one, because a single week can be distorted by a holiday or a launch. Then divide the remaining backlog by the average to estimate the time to clear it.

1. 1

   Choose the period

   Fix a window such as one week or one sprint, and use the same window every time you measure.

2. 2

   Count completed items

   Count only cards that reached the done state and stayed there, excluding reopened or split cards.

3. 3

   Divide by period length

   Express the count as items per week so periods of different lengths can be compared.

4. 4

   Average and forecast

   Take the rolling average across recent periods, then divide the backlog by it to forecast completion.

Velocity dropping is a symptom, not a cause. A metric tree breaks the rate into the factors that produce it: how much work starts, how fast each item moves once started, and how often work gets blocked or reopened. With the tree in front of you, a falling velocity points at a specific branch instead of a general feeling that the team is slow.

KPI Tree connects each branch to the people who own it through RACI, so accountability is explicit rather than assumed. When velocity falls because the blocked rate has climbed, the person accountable for unblocking work is informed directly. This is the difference between a dashboard that shows a decline and a system that gets someone to act on it.

Absolute velocity is meaningless across teams because card sizes and definitions of done vary. What matters is stability over time. A useful benchmark is how much your own velocity varies period to period, since a stable rate forecasts well and a volatile one does not.

Faster velocity comes from smoother flow, not from pushing people harder. The biggest gains usually come from reducing the time cards sit waiting and from making cards small enough to finish in a predictable window.

1. 1

   Comparing teams by raw velocity

   Two teams with different card sizes cannot be compared by count. Velocity is for tracking one board over time, not ranking teams.

2. 2

   Treating velocity as a target

   When velocity becomes a goal, people inflate it by splitting cards or marking work done early. Track it, do not chase it.

3. 3

   Counting reopened work

   A card finished, reopened, and finished again is one piece of work. Counting it twice overstates velocity and hides rework.

4. 4

   Ignoring intake

   A board can have high velocity and a growing backlog if new work arrives faster than it is finished. Always read velocity against intake.