Maintenance work

What it means

Maintenance work is a developer productivity concept that helps teams understand maintenance work in the context of software delivery. Forecasts and comparisons depend on assumptions about history, staffing, dependencies, and scope. It is most useful when tied to a specific work boundary, time period, and decision. The name alone does not tell a team whether a change is good, so interpretation should include the surrounding workflow and the result the team intended to create.

Example

For example, a team examining maintenance work might compare a normal delivery week with a week dominated by a migration, incident, or dependency change. The team records what happened, checks related quality and flow signals, and uses the result to choose one improvement rather than assigning blame. This keeps the concept connected to actual engineering work instead of treating a dashboard value as self-explanatory. A team can also ask whether the signal changed because of the intervention or because the work mix, staffing, release policy, or instrumentation changed.

Limits and interpretation

The main limitation is that maintenance work is a contextual signal, not a complete measure of developer value. A range is often more truthful than one promise; update it when evidence changes. Different roles, work types, and system constraints can produce different results, so comparisons require care. Use the concept to support a concrete improvement question, preserve the definition used, and compare like with like. If the evidence is incomplete, say so and invite the people doing the work to explain what the data cannot show.

How it fits with Weave

Weave's engineering intelligence can help teams investigate maintenance work alongside delivery, quality, review, and developer experience signals. Those signals can reveal patterns and help frame a useful conversation, but they do not establish individual value or replace product context, qualitative evidence, or judgment about the system.