The ICE Scoring Model for Prioritization Explained
ICE is a fast, simple prioritization method that scores each idea on three factors — Impact, Confidence, and Ease — each rated on a simple scale (commonly 1-10), then multiplied together for a single score. It's designed to be quicker and less data-intensive than more detailed frameworks like RICE, trading some precision for speed — useful when you need a reasonable gut-check ranking across many ideas quickly, not a rigorous, data-backed analysis.
Quick facts
- The formula is: ICE score = Impact × Confidence × Ease
- Each factor is typically scored on a 1-10 scale, based on team judgment rather than hard data.
- ICE is faster to run than RICE, since it skips the separate Reach calculation, but is also less rigorous.
- It's best used for quick, early-stage prioritization or for comparing many rough ideas fast, not for high-stakes, well-resourced decisions.
- Created by Sean Ellis, a growth marketing expert, originally for prioritizing growth experiments.
The three factors, explained
| Factor | What it measures | How to score it |
|---|---|---|
| Impact | How much this idea would help if it works | 1 (minimal) to 10 (massive) |
| Confidence | How sure you are this will actually deliver that impact | 1 (pure guess) to 10 (very confident, backed by real evidence) |
| Ease | How simple this is to actually implement | 1 (very difficult) to 10 (very easy) |
A worked example
A growth team is comparing three quick experiment ideas:
Idea A — Add a simple onboarding checklist
- Impact: 6, Confidence: 7, Ease: 8
- ICE score = 6 × 7 × 8 = 336
Idea B — Rebuild the entire referral program from scratch
- Impact: 9, Confidence: 5, Ease: 2
- ICE score = 9 × 5 × 2 = 90
Idea C — Add a small pricing page tweak
- Impact: 3, Confidence: 8, Ease: 9
- ICE score = 3 × 8 × 9 = 216
Idea A scores highest — a solid balance of decent impact, reasonable confidence, and high ease. Idea B has the highest potential impact, but its low ease and moderate confidence pull its score down significantly — useful for flagging that while it might be the biggest bet, it's also the riskiest and most resource-intensive one, worth a separate, more careful conversation before committing.
Why ICE trades precision for speed
Unlike RICE, which requires an actual Reach number based on real usage or market data, ICE relies purely on team judgment across all three factors — this makes it much faster to run (useful for scoring a long list of ideas quickly) but also more subjective and prone to bias, since there's no real data forcing more rigorous, evidence-based scoring. ICE works best as an early filtering tool, or for lower-stakes decisions where speed matters more than precision — for major, resource-intensive commitments, a more rigorous framework like RICE is usually worth the extra time.
ICE vs RICE
| ICE | RICE | |
|---|---|---|
| Factors | Impact, Confidence, Ease | Reach, Impact, Confidence, Effort |
| Speed | Fast — pure judgment-based scoring | Slower — requires real Reach and Effort estimates |
| Precision | Lower — more subjective | Higher — grounded in more concrete numbers |
| Best for | Quick filtering, growth experiments, early-stage ideas | Larger, higher-stakes prioritization decisions |
Many teams use ICE as a fast first-pass filter across a large list of ideas, then apply RICE (or a similarly more rigorous process) to the smaller shortlist of ideas that survive that first filter.
Common mistakes when using ICE
- Treating ICE scores as precise, objective numbers, when they're inherently subjective, judgment-based estimates — useful for rough relative ranking, not exact comparison.
- Having one person score all three factors alone, rather than as a team discussion. The conversation that happens while scoring together often reveals more useful insight than the final number itself.
- Using ICE for major, resource-intensive decisions where more rigor is actually warranted. ICE's speed is a feature for the right use case, but a real limitation for bigger bets.
- Not recalibrating the scoring scale across the team. If one person's "7" means something different from another's "7," the resulting scores aren't genuinely comparable.
FAQ
Who created the ICE scoring model? Sean Ellis, a growth marketing expert credited with popularizing the term "growth hacking," created ICE originally as a way to quickly prioritize growth experiments.
Is ICE only used for growth experiments, or does it work for general feature prioritization? While it originated in growth marketing, ICE is now used broadly for general feature and initiative prioritization whenever a fast, lightweight ranking method is more valuable than a slower, more data-intensive one.
What's a "good" ICE score? Like RICE, there's no universal good score — ICE scores are only meaningful relative to the other ideas being compared in the same session, since the underlying 1-10 scales are subjective and specific to that team's calibration.
Should a team switch from ICE to a more rigorous framework as it grows? Often yes — as more resources and higher stakes get attached to prioritization decisions, many teams shift toward more rigorous, data-backed frameworks like RICE for their major decisions, while still using ICE for fast, lower-stakes filtering.