Software Architecture Metrics

Chapter 10. Measure the Unknown with the Goal-Question-Metric Approach

Michael Keeling

In software, as in life, the most important things are often the most challenging to measure. How much technical debt is in the system, and where should you invest? How well does the architecture, as implemented, satisfy the most important quality attributes? How is the team’s design maturity progressing? For big questions like these, it’s easy to venture a guess based on gut feelings, which can be biased and unreliable. It’s far better to use data. When I need to define metrics for gnarly, difficult-to-answer problems like these, I turn to the goal-question-metric (GQM) approach.

The GQM approach is an analysis technique proposed by Victor Basili and David Weiss to help teams figure out how to measure and evaluate tough problems in software development.¹ The technique is easy to learn and straightforward to apply. It can be used alone or as part of a collaborative workshop. Discovering the right metrics to evaluate a difficult problem still requires creativity and analytical thinking, but GQM provides just enough structure to nudge teams toward better outcomes.

In this chapter, you will learn how to use the GQM approach and facilitate a collaborative workshop with GQM at its heart. You’ll also see a case study about how a software development team used GQM to evaluate and improve gaps in their architecture. By the end of this chapter, you will know everything you need to apply GQM with your ...