Why Scaling Agile Doesn't Work
summary
Scaling agile fails because teams optimize iteration speed while the portfolio and funding cycle keep lead time huge: 47% committee decisions vs 24% economic frameworks, so big projects get funded and wrong ideas grind on. Case studies show mesque has three features costing about $2–2.8M per week of delay, Amazon deploys every 11.6 seconds on average, and HP LaserJet shifts 400 developers to trunk with a CI pipeline, 30,000 test hours, removes a six‑week stabilization phase and achieves roughly 8x productivity toward innovation.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:25The problem with water-scrum-fall
- 3:56Flawed decision-making
- 5:58The high cost of failed projects
- 8:24Estimation versus value
- 12:18Cost of delay
- 14:33Shifting to an experimental approach
- 17:11Impact mapping and outcomes
- 20:58Hypothesis-driven delivery
- 23:46Software's unique advantages
- 27:58HP LaserJet case study
- 32:45Building quality in
- 35:41Measuring transformation success
- 38:41The Improvement Kata
- 42:00Setting target conditions
- 44:12Conclusion
- 45:28Q&A
related talks
39mScrum at Scale
58mScaling Agile in Organizations
26mAgile Transformations are not about software
48mTwo Dimensions Of Agile At Scale
58mAgility at Scale: Tactical and Strategic Approaches
45m"Moving Fast at Scale"
29mEconomies of Speed
52mSmall Steps Are the Fastest Way Forward: Life Beyond Agile & Scrum