Kanban guide · Flow and metrics
Throughput in Kanban: How to Measure and Use It
Throughput is the number of work items a team finishes in a given period, such as a week. It is simple to measure and, combined with cycle time, is the basis of Kanban forecasting.
Key takeaways
- Throughput is items finished per period.
- It powers simple and Monte Carlo forecasts.
- Raise it by removing bottlenecks and rework, not by starting more.
Measuring throughput
Count the items that crossed your delivery point in each period. Plot the counts as a run chart, week by week, to see variation and trends. Unlike velocity, throughput counts items, not story points, so no estimation is needed.
Throughput and Little's Law
Average throughput = average WIP ÷ average cycle time. For example, a team with 12 items in progress and an average cycle time of 3 weeks finishes about 4 items a week. Adding more WIP does not raise throughput once the team is at capacity; it only lengthens cycle time.
Forecasting with throughput
- Simple: 60 items remaining ÷ about 6 per week ≈ 10 weeks.
- Range: use the slowest and fastest recent weeks for a pessimistic and optimistic forecast.
- Monte Carlo: randomly sample past weekly throughput thousands of times to get probabilities, such as "85% chance of finishing by 30 May".
Improving throughput
- Remove bottlenecks; see Theory of Constraints.
- Reduce rework by improving quality early.
- Cut time spent blocked; see blocker clustering.
- Automate repetitive steps.
- Keep WIP limited so work finishes rather than piling up.
Free templates for this topic
Frequently asked questions
What is the difference between throughput and velocity?
Throughput counts finished items per period; velocity sums story points per sprint. Throughput needs no estimation.
Should throughput keep going up?
Not necessarily. Stable, predictable throughput with short cycle times is usually more valuable than a rising number.
What is throughput in Kanban?
Throughput is the number of work items finished in a period, such as items per week.
How do you calculate throughput?
Count the items that reached Done in each period. For example, 6, 8 and 7 items over three weeks gives an average throughput of 7 per week.
How is throughput used for forecasting?
Divide remaining items by typical throughput for a rough date, or run a Monte Carlo simulation on past weekly throughput for a probability-based forecast. See Kanban planning.
How does throughput relate to cycle time?
By Little's Law, average cycle time = average WIP ÷ average throughput. At steady throughput, reducing WIP shortens cycle time.
Should all items count equally in throughput?
Many teams count items regardless of size, which works well if items are split to a similar size. Track throughput by work type if items differ greatly.
What is a throughput run chart?
A chart showing throughput per period over time, helping you spot trends, seasonality and the effect of process changes.
How can a team increase throughput?
Relieve the bottleneck, reduce blockers and rework, and limit WIP so effort goes into finishing. See Theory of Constraints.
Why does throughput vary week to week?
Items differ in size, holidays and absences change capacity, and blockers come and go. Use ranges and percentiles rather than a single number.
Is throughput a productivity measure?
It measures delivery rate, not individual productivity. Use it to understand and forecast the system, not to compare people or teams.