Back to Blog
Implementation Guide

How to Prioritize Experiments Using ICE Framework

Updated December 2025
10 min read
TL;DR

ICE Framework: Score each experiment on Impact (1-10), Confidence (1-10), and Ease (1-10). Average the scores. Run highest-scoring experiments first. This prevents wasting time on low-impact tests or getting stuck on hard-to-implement ideas.

What is ICE?

I = Impact (1-10)

How much will this move the needle if it works? A checkout redesign has higher impact than a button color change.

C = Confidence (1-10)

How confident are you this will work? Based on data, research, or past experiments. Wild guesses = low confidence.

E = Ease (1-10)

How easy is this to implement? A copy change is easier than a full redesign. Consider dev time, design needs, and risk.

How to Calculate

ICE Score = (Impact + Confidence + Ease) / 3

Example:
Impact: 8, Confidence: 6, Ease: 7
ICE Score = (8 + 6 + 7) / 3 = 7.0

Example Prioritization

ExperimentImpactConfidenceEaseScore
Add trust badges5897.3
Simplify pricing page8666.7
Redesign checkout flow9736.3
Change CTA button color24105.3

In this example, "Add trust badges" wins because it's high confidence, easy to implement, and moderate impact.

Tips for Better Scoring

  • Be honest about confidence: Don't inflate scores for pet ideas
  • Score as a team: Average scores from multiple people
  • Revisit scores: Update based on new data or learnings
  • Document reasoning: Note why you scored each dimension

Start Prioritizing

Use ICE to build your experiment backlog, then run tests with ExperimentHQ. Our visual editor makes "Ease" scores higher for most experiments.

Share this article

Ready to start A/B testing?

Free forever plan available. No credit card required.