Throughput variability
Also known as Delivery rate variability, Completion variability
Definition
Throughput variability is the spread of completed-item counts across time periods for a defined work population. It affects capacity planning and forecasting because the average rate does not describe every interval.
The average rate is not every week
Throughput variability is normal in knowledge work. Work size, incidents, leave, dependencies, and release policies can change how many items complete in a period. Forecasting with only the average hides the risk that a future interval will land above or below it.
A concrete example
An illustrative team completes 5, 8, 7, and 12 comparable items across four weeks. Its average is eight, but the observed range is five to twelve. A forecast should communicate that variation and explain whether the work mix is changing.
Reduce avoidable variation
Use smaller coherent items, limit WIP, separate service classes, and identify recurring constraints. Do not remove difficult intervals from the history simply because they make planning uncomfortable. Investigate before smoothing.
Limitations
Short samples are unstable, and item counts hide scope. A planned release batch can create legitimate variation. State the interval, population, and treatment of exceptional work.
How this relates to Weave
Weave can help teams examine completed engineering changes across time and relate unusual intervals to delivery context. Use that history to express ranges, not promises, and include work that the repository view does not contain.
Explore Engineering intelligence