01
Minimum Viable Product
An MVP is not necessarily a low-quality product. It is the smallest implementation capable of testing the most important business or product hypothesis with real users.
02
The 90/10 Principle
Early-stage startups should aggressively search for high-leverage solutions rather than automatically building the most complete engineering system.
03
Iteration Speed
learning velocity ≈ validated decisions / unit time
The purpose of speed is not activity. Speed matters because faster experimentation can reduce uncertainty sooner.