Most projects fail because they answer the wrong question beautifully.
Before a line of production code is written, we want one thing in front of real people. Not a deck, not a prototype that only survives a happy path — something they can use badly and still teach us something.
Scope is not about doing less. It is about arranging the work so the riskiest assumption is tested first, while changing course is still cheap.