
Optimization Through Feedback Loops

Unbundling the Enterprise: the Power of APIs, Optionality, and the Science of Happy Accidents
Topic 8 of 10
Published Oct 21, 2024 · Clip 38:29–45:24 (6:56)
The authors explain optimization through feedback loops, opening with the story of Marvin Pipkin, who spent 20 years cataloging failed experiments before accidentally inventing shatterproof light bulbs. Lowering the cost of experimentation via feature flags, ramps, statistical literacy, and automated feedback loops enables wide optionality—illustrated by the two pirates metaphor (one map vs. a thousand shovels). APIs themselves are instrumentation that drives observability and feedback. They close citing Mik Kersten's insight that transformations framed in technology goals rather than business goals are doomed to fail.