Score every idea. Ship the right one.

ICE was popularized by growth-hacker Sean Ellis as a fast way to triage a long backlog without arguing for an hour. Three axes:

• Impact — how much does this move the needle if it works?
• Confidence — how sure are we it'll work? (Past data > hunch.)
• Ease — how cheap is it to ship? (Engineering + ops + risk.)

Use 1-10 (not 1-5) so the spread is wide enough to actually rank things. Score = I × C × E (max 1000). Tied scores? Tiebreak on Ease — cheap wins.

RICE is the same idea with one more axis (Reach). Use RICE when you have real reach data; use ICE when you don't and want to ship before the meeting ends.