Conversion Rate Calculator
Work out conversion rate — or find out whether variant B actually beat variant A, with a real significance test built in, not just two percentages side by side.
Mode
Variant A — control
Variant B — challenger
Nothing is sent anywhere. Every calculation runs in your browser, and no figure you type leaves this page.
How to use this calculator
- Solve conversion rate when you have visitors and conversions and want the percentage — the standard after-the-fact calculation.
- Solve conversions when you're forecasting: "at a 3% conversion rate, what does 20,000 visitors produce?"
- Solve visitors needed when you're working backwards from a goal: "I need 400 conversions at a 2.5% rate — how much traffic do I need?"
- Compare two variants when you're about to declare an A/B test winner. Enter both, and the tool tells you how confident you can be that the difference is real, not noise.
Why the compare mode exists
Here is the mistake almost every CRO test makes.
A landing page test runs for four days. Variant A converts at 2.0%. Variant B converts at 2.7%. Someone declares B the winner and ships it everywhere.
But if each variant only saw a few hundred visitors, that gap can easily be inside the range pure chance produces. Ship the "winner" on that evidence and you may have just rolled out a change that does nothing — or actively hurts conversion — based on noise.
The compare mode runs a two-proportion significance test on your actual numbers and returns a confidence level. Above 95%, the difference is unlikely to be chance. Below 90%, you do not yet know anything and the test needs to keep running. The full methodology behind this — sample size planning, when to stop a test, common testing mistakes — is covered in our A/B testing guide; this tool runs the maths on the numbers you already have.
The formulas
| You want | Formula |
|---|---|
| Conversion rate | (conversions ÷ visitors) × 100 |
| Conversions | visitors × (rate ÷ 100) |
| Visitors needed | conversions ÷ (rate ÷ 100) |
| Significance | two-proportion z-test on pooled standard error |
The significance maths is written out in full in the comment at the top of this page's source, so anyone can check it rather than take our word for it. The full explanation of the metric itself is on our conversion rate guide, and the complete CRO playbook — landing page design, cart abandonment, pricing pages, funnels and more — starts at our Conversion Rate Optimization guide.
Where a small lift in conversion rate matters more than people think
- It compounds with traffic, not against it. A 0.5-point lift on a page that already gets meaningful traffic is worth more than the same lift on a page nobody visits — see how to increase conversion rate for where to look first.
- It is the cheapest lever in performance marketing. Improving conversion rate lowers your effective cost per acquisition without touching ad spend at all.
- Where visitors actually leave. A rising or falling conversion rate is a symptom — the conversion funnel guide is where to diagnose the specific step losing people.
- Landing page fundamentals move it fastest. Before running any test, check the page against our landing page design principles.
Why we do not publish a "good conversion rate"
Because the number is close to meaningless without industry, traffic source, price point and funnel-stage context.
A 2% conversion rate is strong for cold paid traffic to an ecommerce store and poor for a warm email list clicking through to a low-price offer. Any single benchmark is averaging across situations that do not compare.
Your own account history is the real benchmark. And inside a test, the other variant is the benchmark — which is exactly what the Compare mode above is for.
Frequently asked questions
How do you calculate conversion rate?
Divide total conversions by total visitors and multiply by 100 to get a percentage. 250 conversions from 10,000 visitors works out to a 2.5% conversion rate. The same formula applies whether you're counting purchases, signups, form fills or any other action you define as a conversion.
What's a good conversion rate in India?
There's no single defensible number — conversion rate varies hugely by industry, traffic source, price point and funnel stage. A 2% rate is strong for cold paid traffic to an ecommerce store and poor for a warm list clicking a low-price offer. Compare against your own historical data, not a generic benchmark.
How do I know if my A/B test result is real?
Use the Compare mode on this page. It runs a two-proportion significance test on your actual visitor and conversion counts and returns a confidence level. Below 90% confidence, treat the "winner" as unproven — the gap you're seeing may just be normal random variation, and shipping a change on that basis is a coin flip.
Is a higher conversion rate always better?
Not automatically. Conversion rate only measures how often visitors convert — it says nothing about what that conversion is worth. A variant that converts more people into lower-value customers can produce less total revenue than one that converts fewer, higher-value ones. Always check average order value and downstream revenue before fully committing to a result.
Is my data sent anywhere when I use this calculator?
No. Everything is calculated in your browser as you type. Nothing you enter is sent to us or stored — refresh the page and it's gone.
Numbers look better. Are they actually significant?
Send us your test data — visitors, conversions, both variants. We'll tell you honestly whether it's a real result or a coin flip.