Flow and capacity planning

Capacity planning

Also known as Engineering capacity planning, Team capacity planning

By WeavePublished 1 min read

Definition

Capacity planning is the practice of estimating the work a delivery system can complete and comparing it with expected demand, constraints, and service commitments. It supports tradeoffs rather than promising exact output.

A decision aid, not a promise

Capacity planning asks what a system can likely handle under stated conditions. It should account for planned work, support, incidents, leave, dependencies, and variation. A plan that fills every nominal hour often leaves no room for the uncertainty that creates queues.

A concrete example

An illustrative team historically completes between 8 and 12 similarly sized items per month. Its next month includes a known migration and an on-call rotation. Rather than promise 12 feature items, the team reserves capacity for the migration and expresses the remaining forecast as a range.

Plan around constraints

Compare arrival rate, throughput, WIP, and blocked time. If review capability limits delivery, adding implementation capacity may increase unfinished work. If demand exceeds capacity, make the tradeoff visible by changing scope, sequence, service expectations, or staffing.

Limitations

Historical throughput is not a guarantee, especially after architecture, team composition, or work mix changes. Story points also do not create a common unit across teams. State the population, horizon, and uncertainty behind every forecast.

How this relates to Weave

Weave can provide historical engineering activity and delivery patterns that help teams ground a capacity conversation. It does not forecast every source of demand or decide priorities, so planning should combine analytics with product, operational, and people context.

Explore Engineering intelligence

Sources and further reading

  1. The Kanban Guide