Minimum detectable effect (MDE) calculator

Use this calculator when your outcome is binary — for example, converted or not, signed up or not, or active on day 7 — and your test's total sample size is already fixed or limited, rather than something you are free to choose.

If you already know the effect you want to detect and need to find the required number of users instead, use the sample size calculator.

• The test duration is fixed, for example a one-week campaign or a seasonal promotion.
• Traffic is limited, for example on a low-traffic page, a niche segment or a B2B product.
• You want to check, before launching, whether an A/B test is worth running or is likely to be underpowered.
• You need to set realistic expectations with stakeholders about which effects the test can and cannot detect.
• You are deciding whether to increase traffic, extend the test duration or test a bolder change.

The calculation assumes:

• Random assignment: observations are independently assigned to the two groups.
• Independent observations: each randomized unit contributes one observation, and observations from different units are assumed to be independent.
• A sufficiently large sample: the calculation uses a normal approximation to the binomial. For very small expected counts, treat the result as approximate.

The detectable lift is the smallest relative improvement your A/B test can reliably detect with the available sample size, and the absolute difference shows the same effect in absolute terms (percentage points for conversion rates, or the metric's units for continuous KPIs).

If the true effect is at least as large as the minimum detectable effect, your test has a probability of at least the selected statistical power (for example, 80%) to produce a statistically significant result. Smaller effects can still reach statistical significance, but with a lower probability.

Compare the detectable lift with the effect you realistically expect from the change. If the detectable lift is larger than the expected effect, the test is underpowered and is likely to miss a real improvement. In that case, consider increasing the sample size, running the test longer, or testing a bolder change.

Suppose your baseline conversion rate is 10% and your test can reach a total of 7,682 users, split evenly.

If the true effect is at least a 20.00% lift, this sample size gives you an 80% probability of detecting it as statistically significant. A smaller true effect would still be worth having, but this test is not powered to reliably catch it.

Use this calculator when your outcome is continuous — for example, revenue per user, session duration, or time on page — and your test's total sample size is already fixed or limited, rather than something you are free to choose.

If you already know the effect you want to detect and need to find the required number of users instead, use the sample size calculator.

• The test duration is fixed, for example a one-week campaign or a seasonal promotion.
• Traffic is limited, for example on a low-traffic page, a niche segment or a B2B product.
• You want to check, before launching, whether an A/B test is worth running or is likely to be underpowered.
• You need to set realistic expectations with stakeholders about which effects the test can and cannot detect.
• You are deciding whether to increase traffic, extend the test duration or test a bolder change.

The calculation assumes:

• Random assignment: observations are independently assigned to the two groups.
• Independent observations: each randomized unit contributes one observation, and observations from different units are assumed to be independent.
• Approximately normal distribution: the calculation relies on the sampling distribution of the mean being approximately normal, which holds well for reasonably large samples even if the underlying metric is skewed.
• Equal variance between groups: this calculator's formula uses the pooled two-sample t-distribution (degrees of freedom = n₁ + n₂ − 2), which assumes both groups share the standard deviation you enter. If the two groups are likely to have materially different spreads, treat the result as approximate.

The detectable lift is the smallest relative improvement your A/B test can reliably detect with the available sample size, and the absolute difference shows the same effect in absolute terms (percentage points for conversion rates, or the metric's units for continuous KPIs).

If the true effect is at least as large as the minimum detectable effect, your test has a probability of at least the selected statistical power (for example, 80%) to produce a statistically significant result. Smaller effects can still reach statistical significance, but with a lower probability.

Compare the detectable lift with the effect you realistically expect from the change. If the detectable lift is larger than the expected effect, the test is underpowered and is likely to miss a real improvement. In that case, consider increasing the sample size, running the test longer, or testing a bolder change.

Suppose your baseline mean is 10, with a standard deviation of 1, and your test can reach a total of 128 users, split evenly.

If the true effect is at least a 5.00% lift, this sample size gives you an 80% probability of detecting it as statistically significant. A smaller true effect would still be worth having, but this test is not powered to reliably catch it.

Use this calculator when your outcome is a ratio metric — a value calculated as one quantity divided by another, measured per randomization unit — for example, revenue per user, clicks per impression, or items purchased per session — and your test's total sample size is already fixed or limited, rather than something you are free to choose.

If you already know the effect you want to detect and need to find the required number of users instead, use the sample size calculator.

• The test duration is fixed, for example a one-week campaign or a seasonal promotion.
• Traffic is limited, for example on a low-traffic page, a niche segment or a B2B product.
• You want to check, before launching, whether an A/B test is worth running or is likely to be underpowered.
• You need to set realistic expectations with stakeholders about which effects the test can and cannot detect.
• You are deciding whether to increase traffic, extend the test duration or test a bolder change.

The calculation assumes:

• Random assignment: observations are independently assigned to the two groups.
• Independent observations: each randomization unit contributes its own numerator/denominator pair, and different units are assumed to be independent of each other.
• Approximately normal distribution: the calculation relies on the delta-method approximation for the ratio's variance, which is well-behaved for reasonably large samples.

The detectable lift is the smallest relative improvement your A/B test can reliably detect with the available sample size, and the absolute difference shows the same effect in absolute terms (percentage points for conversion rates, or the metric's units for continuous KPIs).

If the true effect is at least as large as the minimum detectable effect, your test has a probability of at least the selected statistical power (for example, 80%) to produce a statistically significant result. Smaller effects can still reach statistical significance, but with a lower probability.

Compare the detectable lift with the effect you realistically expect from the change. If the detectable lift is larger than the expected effect, the test is underpowered and is likely to miss a real improvement. In that case, consider increasing the sample size, running the test longer, or testing a bolder change.

Suppose your baseline revenue-per-session ratio is $25, with a standard deviation of 3, and your test can reach a total of 46 users, split evenly.

If the true effect is at least a 9.91% lift, this sample size gives you an 80% probability of detecting it as statistically significant. A smaller true effect would still be worth having, but this test is not powered to reliably catch it.