A/B Test Prioritization

10 pre-loaded test ideas scored with the ICE framework. Adjust the scores to match your store, add your own ideas, and build a testing roadmap.

Not every test is worth running. The ICE framework helps you compare ideas objectively so you spend time on the tests most likely to move revenue.

Impact (1–10)

How much will this move the needle on conversions or revenue?

Confidence (1–10)

How sure are you it will work, based on data or precedent?

Ease (1–10)

How easy is it to implement? Higher = less effort required.

ICE Score = (Impact + Confidence + Ease) / 3. Higher score = do it first.

Score Your Test Ideas

Start here
ICE Score8.0
Start here
ICE Score7.7
Start here
ICE Score7.0
ICE Score6.7
ICE Score6.7
ICE Score6.7
ICE Score6.3
ICE Score6.3
ICE Score6.0
ICE Score5.7

Your Top 3 Tests to Run Next

1

Add trust badges above the fold on product pages

ICE 8.0 — Impact 7, Confidence 8, Ease 9 · Quick win — low effort

2

Move reviews/ratings above the fold

ICE 7.7 — Impact 6, Confidence 8, Ease 9 · Quick win — low effort

3

Add free shipping threshold progress bar to cart

ICE 7.0 — Impact 8, Confidence 7, Ease 6 · Moderate effort

Run these in order. Start with the highest-scored test, measure the result, then move to the next. One test at a time gives you clean data.

Frequently Asked Questions

What is the ICE scoring framework?

ICE stands for Impact, Confidence, and Ease. Each test idea is scored from 1 to 10 on all three dimensions, and the average gives you a single prioritization score. It was popularized by Sean Ellis as a lightweight way to compare growth experiments without overthinking the decision.

How many A/B tests should I run at once?

One test per page at a time. Running multiple tests on the same page creates interaction effects that corrupt your results. If you want to move faster, test on different pages simultaneously — for example, one test on the product page and one on the cart page.

How long should I run each test?

Run until you hit statistical significance (typically 95% confidence) and at least one full business cycle — usually 2 to 4 weeks. Ending early because the numbers look good leads to false positives. Tools like Google Optimize and VWO have built-in significance calculators.

What if two tests have the same ICE score?

Break ties by choosing the test with the higher Impact score — a big win with lower confidence is usually more valuable than a small certain win. If Impact is also tied, go with the higher Ease score so you can learn faster and move on to the next test.

More Free Resources

Find out exactly where your store is leaking revenue.

Answer a quick set of multiple-choice questions and we'll pinpoint your biggest revenue leaks — and whether we can help plug them.

Find Your Revenue Leaks

Free · No obligation · 2 minutes