Build, measure, learn. Validated learning rather than progress against a plan. The minimum viable product as an experiment rather than a cheap version of the real thing.
Read April 2011, when this was the dominant vocabulary in product work, and the vocabulary is most of what survived. MVP in particular got absorbed and inverted, and now mostly means a thin first release with the learning part removed.
The pivot material is the part I still use. Deciding in advance what result would make you change direction is harder and more valuable than any of the measurement advice.
