Kill criteria for ideas

Most ideas do not need a decision. They need a check that says no fast enough to be cheap.

The four checks

1. Who asked? Name a person who wanted this. Not a market, not a segment. A person. If the name is hard to produce, the demand is hypothetical.
2. What breaks if we skip it? If nothing breaks, this is a preference, not a problem. Preferences go in the vault.
3. What is the smallest version that proves it? If the smallest version is still weeks of work, the idea is not understood well enough to start.
4. What does it cost to be wrong? Cheap and reversible, just try it. Expensive or hard to undo, it needs a real decision, not enthusiasm.

Using it

Run the checks in order and stop at the first failure. The point is to stop early, so working through all four on a dead idea wastes the tool.

Write down which check killed it. A list of dead ideas with reasons is worth more later than the ideas were.