The End Game
summary
A claim that most teams still run 65% UI end-to-end tests comes with specifics: years of automating with Selenium and TestComplete, cranking out six to eight UI scripts a day until builds break repeatedly, and a LinkedIn poll that confirms the pattern. Concrete counters include the TRIMS and SACRED checklists, a focus on lowering the test layer, a target metric of mean time to feedback (aiming for ~15 minutes), and admission of eight deployed tests that never fail — illustrating what gets built and what actually works.
AI-written summary · about
chapters (7)
- 0:00Automation to testing awakening
- 4:44Brittle end-to-end test trap
- 13:10Test automation lost its way
- 22:21Building automation with system knowledge
- 33:17Designing targeted automated tests
- 42:23Endgame of UI test obsession
- 56:08Q&A
related talks
27mTest Effective Development - End 2 (What) End?
21mLowering End to End Testing Barriers
29mEnd to End Test Automation for Both Horizontal and Vertical Scale
33mDevs should love your tests
27mThe Effective Development - End (What) End
45mTesting the web without testers — success or failure
42mThe Holy Trinity of UI Testing
57mInverting Test Pyramid - A First Hand Experience Report