Developer productivity

Engineering work mix

Also known as Software engineering work mix, Engineering allocation mix

By WeavePublished 2 min read

Definition

Engineering work mix is the distribution of engineering effort or completed work across categories such as features, maintenance, defects, incidents, security, and enablement. It makes competing demands on a software factory visible so capacity and delivery results can be interpreted together.

Capacity is spent on more than features

An engineering team may be asked to deliver a product feature while also maintaining dependencies, responding to incidents, fixing defects, meeting security requirements, and improving the platform. If reports count only planned feature work, they can make the team appear to have unused capacity when the rest of the factory is absorbing demand.

Work categories should reflect the decisions the team needs to make. A small set of stable categories is easier to use than a taxonomy that requires every change to fit a precise label. Document how mixed changes are handled and whether the unit is a pull request, work item, deployment, or time allocation.

Read the mix with outcomes

The percentage of work in a category is descriptive. A rise in incident work may indicate a reliability problem, a temporary event, or better classification. A rise in maintenance can reduce future delivery risk. Compare the mix with cycle time, rework, reliability, product progress, and developer experience rather than declaring one category good or bad.

How Weave can help

Weave can provide a consistent view of code changes and their surrounding workflow. Teams can inspect the examples behind a work-mix trend and ask whether the factory is creating new value, paying down risk, or recovering from failures. Product planning and incident systems add the context needed for work that never becomes a pull request.

How this relates to Weave

Weave can help classify and compare the work represented in pull requests, including code output, review, quality, and delivery context. That gives teams a way to see whether a factory's capacity shifted toward features, maintenance, defects, or recovery work. Classification is an analytical aid and should be checked against team and product context.

Explore Engineering intelligence

Sources and further reading

  1. The SPACE of Developer Productivity, Microsoft Research
  2. DORA value stream management for software delivery