Welcome to Bell's Calculators System
Looking for an A/B test sample size calculator? Want to check whether your data has a sample ratio mismatch (SRM)? Or are you ready to analyze your experiment results? Whatever stage you're at, you're in the right place.
A/B test calculators
One experiment, three moments where the statistics matter. Size it before you launch, check the split once traffic is flowing, and read the result when it lands.
Calculators for common tests
Every other test, in one list: the ones that turn up most often, the exact tests for small tables, and the intervals and equivalence questions an ordinary significance test can't answer.
Not sure which one you need?
No worries. Take our short diagnostic and we'll guide you to the right one — a few questions about what you are comparing and how the data was collected.
Bell's Calculators System puts every statistics calculator you need in one place, so there's no more hunting across different sites for a sample size calculator, an A/B testing calculator, or a t-test calculator. Every one of them shares the same consistent format, which makes the whole system easier to use and understand.
Want to go deeper? Most online calculators hand you a number with little insight into how it was calculated. Ours give you the key outputs plus everything you need to understand them: when to use the calculator, its assumptions, and the formula behind it. A worked example turns the theory into practical numbers you can check against your own.
We build measurement systems for a living — see our A/B testing work or the case studies. These tools are the parts of that work that fit on one page.
Frequently asked questions
Are these calculators free?
How do I know which test to use?
Why does every calculator have two tabs?
How do I know the numbers are right?
Can I share a result with someone?
How do I check for a sample ratio mismatch?
A calculator answers one question. We answer the rest.
Sample size is the easy part. Choosing the metric, handling the users who appear in both groups, deciding what to do when the result lands in the awkward middle — that is the work. If you would rather have that designed properly the first time, talk to us.
