What is good enough principle?

Good enough principle is a concept in personal epistemology: The optimal amount of optimization is not infinite — there is a point where you should stop and move on.

Example: A software engineer spends three weeks refining a data processing pipeline. The first day of work cuts execution time from twelve minutes to four minutes — a 67% improvement that unblocks a team of analysts waiting on fresh data every morning. The second day shaves it to two minutes. By the end of the first week, it runs in forty-five seconds. Good enough for any business purpose. But the engineer keeps going. Week two brings it to thirty seconds. Week three, twenty-two seconds. Each day of optimization now yields fractions of a second. Meanwhile, three feature requests sit untouched, two bugs affect paying customers, and a junior developer has been waiting four days for a code review. The pipeline did not need to run in twenty-two seconds. It needed to run in under five minutes. The engineer crossed that threshold on day one. Everything after was optimization theater — visible effort producing invisible value. The cost was not just the engineer's time. It was the opportunity cost of everything that time could have addressed instead.

This concept is part of Phase 29 (Agent Optimization) in the How to Think curriculum, which builds the epistemic infrastructure for agent optimization.