deliberate-practice

Metrics

One way to come up with useful metrics is to loosely get inspired by the simple and straight-forward methodology of Goal-Question-Metric. You start with a list of goals to be achieved, then you devise questions that would help measure over time how close or far you are from each goal, and if you are moving in the desired direction. Then you devise metrics that would answer the questions.

This document lists a few examples of goals, possible questions associated with those goals, and metrics to help on answering the questions.

Improve adoption and quality of peer reviews of source code

Are the development teams understanding better the importance of reviews?

Metrics:

What is still being detected by static analysis of source code?

Metrics:

Improve estimates for software development work (pre-release)

How effective is sprint planning?

Metrics:

When do we think the backlog will be completed?

Metrics:

How effective is backlog refining?

Metrics:

Decrease the cost of manual non-QA work by QA engineers

How much time do QA engineers spend installing and configuring QA environments?

Metrics: