Prioritization & Frameworks

ICE Scoring

A lightweight prioritization framework that scores initiatives by Impact, Confidence, and Ease — producing a simple rank-order without requiring per-feature reach estimates.

What is ICE Scoring?

ICE scoring was popularised by Sean Ellis (who also created the 40% NPS PMF benchmark) as a fast prioritisation tool for growth experiments. It scores each initiative on three dimensions:

FactorQuestionScale
ImpactIf this works, how big is the effect?1–10
ConfidenceHow sure are we it will work?1–10
EaseHow easy is it to implement?1–10
**ICE Score = Impact × Confidence × Ease**

ICE vs. RICE

DimensionICERICE
ReachNot includedIncluded
SpeedFaster — no Reach estimate neededSlower
Best forGrowth experiments, equal-reach featuresBacklog with variable audience sizes
RiskTreats all features as same audienceMore accurate but more effort

When to use ICE

  • Growth experiments where all features target the full user base
  • Quick stack-ranking when you need a decision in < 30 minutes
  • Early-stage products where Reach estimates are unreliable
  • Comparing experiments within a single funnel stage

ICE pitfall

Because Ease is in the formula, ICE systematically favours quick wins over high-impact hard things. Balance this by also reviewing top ICE items against strategic importance.

Frequently asked questions

Should Ease be higher = easier or higher = harder?

Higher = easier in the standard ICE formula. A score of 9 means it's very easy to implement. Some teams invert this to 'Effort' (higher = more effort) and divide instead of multiply — but this changes the formula direction. Stick to the original: higher Ease = simpler implementation = higher score.

How is ICE different from just gut instinct?

ICE makes the assumptions explicit and forces you to separate the three dimensions. It also creates a shared decision record — 'we scored this 7×3×8 = 168 and ranked it #3' — which is more defensible than 'it felt right'. The value is in the conversation the scoring triggers, not the precision of the number.

Apply ICE Scoring to your real product data

PMRead ingests customer feedback, interviews, and Slack threads — and generates PRDs grounded in real evidence.

Try PMRead free →