Free · open · specification-driven
Measure software engineering well.
A working book on choosing metrics that reflect real outcomes rather than activity: the Flow Framework, the SPACE framework, queueing theory, and DORA metrics, code and quality metrics, product and business outcomes, reliability and security, and how generative AI is reshaping what these numbers mean.
Contents
- 1 Foundations of Measurement
- 1.0 Introduction to Part 1: Foundations of Measurement
- 1.1 Why measure software engineering
- 1.2 Goodhart's law and the psychology of metrics
- 1.3 Outcomes over output: choosing what to measure
- 1.4 Metrics governance and ownership
- 1.5 Data sources and instrumentation
- 1.6 Statistical literacy for engineering metrics
- 2 Flow Metrics
- 2.0 Introduction to Part 2: Flow Metrics
- 2.1 The Flow Framework
- 2.2 Flow items: features, defects, risks, and debt
- 2.3 Flow velocity and flow distribution
- 2.4 Flow time and flow load
- 2.5 Flow efficiency and work in process
- 2.6 Cycle time and its components
- 2.7 Queueing theory
- 2.8 Lean value stream metrics
- 2.9 Pull request and code review metrics
- 2.10 The DORA metrics framework
- 3 Developer Experience and the SPACE Framework
- 3.0 Introduction to Part 3: Developer Experience and the SPACE Framework
- 3.1 The SPACE framework
- 3.2 Satisfaction and well-being metrics
- 3.3 Performance metrics and outcome proxies
- 3.4 Activity metrics and their limits
- 3.5 Communication and collaboration metrics
- 3.6 Efficiency and flow: deep work and interruptions
- 3.7 Developer experience surveys and DevEx metrics
- 4 Code and Quality Metrics
- 5 Product and Business Metrics
- 6 Reliability, Operations, and Security Metrics
- 7 Metrics in the Age of AI
- 8 Building a Metrics Program
- 9 Appendices
This site is built with SvelteKit and the Lily Design System. The book's content lives in the software-engineering-metrics content repository; see the project page for how the two fit together, and contributing for how to help.