The Rise and Fall of Feature Teams
summary
Claim: cross‑functional feature teams break down at medium scale and spawn new silos and unowned black boxes. Evidence: a cited study finds 75% dysfunctional setups, internal counts range 200–600 Android engineers, and microservices/mono‑repo choices inflate complexity. Tools tried: feature toggles and release trains, a core team that supplies frameworks (which often becomes a detective squad), and ad‑hoc mobbing/pairing. Outcome: return to platform ownership, hiring a mobile QA after 15 years, and clearer code ownership that actually improves pairing and reliability.
AI-written summary · about
chapters (8)
- 0:00Introduction to team models
- 1:38Origins of feature teams
- 4:25Challenges in team scaling
- 7:17Silos and Conway's Law
- 10:01KPIs and technical debt
- 12:28Managing core infrastructure
- 14:00Refining architecture needs
- 17:37Moving to platform teams
related talks
26mThe Rise and Fall of Feature Teams
33mFeature Teams vs. Platform Pain
46mWhy Cross-Functional Teams Build Better Microservices
01h 10mAgile Cross Team Collaboration How Tos
15mAgility at scale: building feature teams for huge products
45mThe Goldilocks Problem of Software Delivery
20mWorking with Others and Delivering as a Team — Bridges Seen Through Human Relationships
58mDevs fight your Battle! A Pledge for cross-functional Teams