Is the difference real, or just noise?
Enter visitors and conversions for two versions of a page, or for two periods, and get a plain answer with the numbers behind it.
We run this test for you, daily.
Every day GoTrace compares each metric with the same weekday in previous weeks, and only changes that pass a significance test become insights, each with its evidence and confidence.
About this calculator
What does "statistically significant" mean here?
That a difference this large would be unlikely (less than 5% likely at the 95% level) if the two versions really converted at the same rate. It does not tell you the difference is big, only that it is probably not chance.
Which test does the calculator use?
A two-sided two-proportion z-test on the conversion rates, the standard test for A/B tests with a yes/no outcome. The confidence interval is for the difference between the two rates.
Why does it say I need more data?
With fewer than 30 visitors in a group or fewer than 10 conversions in total, the test is not reliable, so the calculator will not call a winner. The sample size shown is how many visitors per version you would need to detect the difference you see, 80% of the time.
Can I compare this week with last week?
Yes. Treat the two periods as A and B. Keep in mind that periods can differ for other reasons (season, campaigns, a holiday), which a test cannot rule out.
Should I stop my test as soon as it is significant?
No. Checking again and again and stopping at the first significant result inflates false positives. Decide the sample size up front and look once it is reached.
More free checks
No sign-up, nothing stored. See all free tools.