Requirement Coverage: Myth or an Ultimate Goal
summary
Calls requirement coverage a metric and shows why a raw percentage (50% vs 100%) misleads, then demonstrates three concrete fixes: use model-based testing to compute complexity, use specification by example and acceptance criteria to count examples as tests, and treat BDD feature files as a single source of truth that syncs to Azure DevOps. A prototype links tests to code and surfaces a coverage heat map in Visual Studio so lines and UI elements show which requirements exercised them; the prototype works.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:33Why requirement coverage
- 4:26Requirement coverage defined
- 9:23Evaluating requirement coverage
- 15:20Improving requirement coverage
- 17:46Model-based testing
- 19:55Specification by example
- 26:31Granularity and acceptance criteria
- 32:32Summary
- 44:48Innovative ideas
- 47:16Q&A
related talks
51mDebunking Myths Around Test Coverage
10m99% is not enough - full test coverage and why it's awesome
39mHow to Talk about Coverage
03mTest coverage
30mMeasuring Coverage and why Cypress Cloud UI Coverage is the Holy Grail
40mAdvanced Test Coverage Criteria: Specify and Measure, Cover and Unmask
49mTest Coverage Myth Busted: Test-to-Code Map and Test Impact Analytics
45mPanel Discussion | How to deal with bad requirements in testing?