🧊ICE scoring

Three numbers for ranking things that compete for the same time. It comes from growth teams, where it is usually credited to Sean Ellis. Every project note in this vault carries the three.

1. Impact. If this works, how much does it matter? 1 to 10.
2. Confidence. How sure am I that it will work, on evidence and not on hope? 1 to 10.
3. Ease. How little will it cost in time and trouble? 10 is an afternoon.
4. Score. The project notes show the average of the three.

Two of mine, as scored on the day

The spare room outscored the most interesting work of the autumn. That is the method working as designed, and a reason not to obey it.

Use when: there are more than five candidates and I am about to pick the one I like.

Not for:

• Comparing a room with a product. The scale is only honest inside one list.
• Anything where Confidence is a guess about a guess. In November ICE put the ranked list first among the scorecard prototypes. The sessions showed it is the part people trust least.