Shopify A/B Test
Significance Calculator

We help 7-9 figure brands scale revenue more efficiently with landing pages + A/B testing

Calculate the statistical significance of your A/B test results.

How the calculator works

1. You Enter Two Key Numbers for Each Version

Visitors: How many people saw Version A, and how many people saw Version B?
Conversions: Out of those visitors, how many took the desired action (like making a purchase)?

2. The Calculator Does the Math

It uses a statistical formula (often based on something called the “Z-test” or a similar approach) to figure out whether the difference in conversion rates is big enough to be more than just luck.
Conversion rate simply means “Conversions divided by Visitors”—for example, if 50 people bought something out of 1,000 visitors, your conversion rate is 5%.

3. It Returns a Confidence Level

This is expressed as a percentage, like 95%. That means there’s only a 5% probability (or less) that you got these test results by sheer random chance.
In other words, if the calculator says 95% confidence, it’s suggesting you can be fairly confident that the difference between Version A and Version B is real, not just random.

4. You Make the Decision

If the confidence is high (commonly 90% or above), many marketers go ahead and pick the winning version (the one with the higher conversion rate).
If the confidence is low, it may mean you need more data or that both versions perform about the same.

1. You Enter Two Key Numbers for Each Version

Visitors: How many people saw Version A, and how many people saw Version B?
Conversions: Out of those visitors, how many took the desired action (like making a purchase)?

2. The Calculator Does the Math

It uses a statistical formula (often based on something called the “Z-test” or a similar approach) to figure out whether the difference in conversion rates is big enough to be more than just luck.
Conversion rate simply means “Conversions divided by Visitors”—for example, if 50 people bought something out of 1,000 visitors, your conversion rate is 5%.

3. It Returns a Confidence Level

This is expressed as a percentage, like 95%. That means there’s only a 5% probability (or less) that you got these test results by sheer random chance.
In other words, if the calculator says 95% confidence, it’s suggesting you can be fairly confident that the difference between Version A and Version B is real, not just random.

4. You Make the Decision

If the confidence is high (commonly 90% or above), many marketers go ahead and pick the winning version (the one with the higher conversion rate).
If the confidence is low, it may mean you need more data or that both versions perform about the same.

1. You Enter Two Key Numbers for Each Version

Visitors: How many people saw Version A, and how many people saw Version B?
Conversions: Out of those visitors, how many took the desired action (like making a purchase)?

2. The Calculator Does the Math

It uses a statistical formula (often based on something called the “Z-test” or a similar approach) to figure out whether the difference in conversion rates is big enough to be more than just luck.
Conversion rate simply means “Conversions divided by Visitors”—for example, if 50 people bought something out of 1,000 visitors, your conversion rate is 5%.

3. It Returns a Confidence Level

This is expressed as a percentage, like 95%. That means there’s only a 5% probability (or less) that you got these test results by sheer random chance.
In other words, if the calculator says 95% confidence, it’s suggesting you can be fairly confident that the difference between Version A and Version B is real, not just random.

4. You Make the Decision

If the confidence is high (commonly 90% or above), many marketers go ahead and pick the winning version (the one with the higher conversion rate).
If the confidence is low, it may mean you need more data or that both versions perform about the same.

Free Resources