Cognitive Scaffold

Preparing your thinking workspace

arrow_back_ios_new
MENTAL MODEL · M6647

Goal-Question-Metric

Goal-Question-Metric
BusinessHigh supportManagement
Included
account_tree

Updated 2026-08-09

Loading revision record…

INTRODUCTION

English translation pending.

CORE DEFINITION

Goal-Question-Metric (GQM) is a measurement framework introduced by Victor Basili and colleagues in the 1980s for software and process measurement. It has three levels. A Goal defines the purpose, the object being measured, the quality focus, and the viewpoint, all at the conceptual level. Questions express what must be known to judge whether the goal is being met, at the operational level. Metrics specify the data to be collected to answer each question, at the quantitative level. The framework's constraint is direction: goals come first, and any metric that cannot be traced back through a question to a goal should not be collected.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

- Use goal statement: write the purpose, object, quality focus, and viewpoint before choosing any metric. - Use question derivation: ask what you would need to know to judge whether the goal is met. - Use metric pruning: delete any metric that no question actually requires.

anchor

Anchor fast decisions

Measurement programs accumulate metrics because they are easy to collect, and then produce reports that nobody uses for decisions. GQM reverses the direction of derivation: the goal determines what questions matter, and the questions determine what data is needed. Because every metric traces back to a stated purpose, the resulting data answers a question someone actually asked, which makes it actionable. The framework also bounds collection cost, since any measurement without a downstream question is cut before it consumes effort.

MINIMUM ACTION

In progress 0/1

Practice this model in one real situation:

Check to track your progress (stored locally)
Learning progress0%
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more

Source support: Explicit

RELATED MODELS