Conversion rate
Conversions divided by visitors for each version. The headline number is how much higher or lower B is than A.
Free tool · Calculator
Did B win, or did it get lucky?
A higher conversion rate after a few hundred visits can be chance. Enter the visitors and conversions for each version and see how confident you can be that the difference is real.
Use the tool
- Free
- No sign-up
- Runs in your browser
Your numbers
–Change in conversion rate, B against A
–Confidence the difference is real
–p-value
–Visitors needed per version
–Still to collect per version
Bulk modeCSV, TSV or paste from a spreadsheet
Upload a list of tests with visitors and conversions for A and B. Every test gets the change, the confidence, the visitors needed and a verdict.
Settings for every rowUsed when a column is missing
Results
Runs in your browser. Nothing you type or upload is stored or sent.
How to read it
Reading the result.
Confidence
How sure you can be that the difference is not chance, from a two-sided test on the two conversion rates. 95% is the usual bar for a decision.
Sample size
The visitors each version needs to detect a difference this size with 80% power. Stopping before that point is how most tests pick the wrong winner.
Run full weeks
Weekdays and weekends convert differently. Run tests for whole weeks, and do not change budgets or offers halfway through.
More free tools
Keep going.
5.60x ROAS
Calculator
Open toolROAS and break-even calculator
See your ROAS, your break-even ROAS and the profit left after ad spend. Bulk CSV mode included.
18%ACoS
TACoS 7.4%Break-even 32%
Calculator
Open toolACoS and TACoS calculator
ACoS, TACoS and break-even ACoS for Amazon sellers. Bulk CSV mode included.
Impressions412K
Clicks9,870
Conversions414
CTR2.4%CPA$36.20
Calculator
Open toolAd metrics calculator
CPM, CPC, CTR, conversion rate, CPA and ROAS from one set of numbers. Bulk CSV mode included.
Questions
Before you ask.
Can I do many at once?
Yes. Switch to Bulk mode, upload or paste a CSV with a header row and map the columns. Every row is processed in your browser and you download the results as a CSV. Nothing is uploaded to us.
Can I stop the test as soon as it reaches 95%?
Checking every day and stopping at the first good result inflates false winners. Decide the sample size first, using the visitors needed figure, and judge the test when you reach it.
Does this work for ads as well as pages?
Yes. Use clicks or impressions as visitors and conversions as conversions, for two ads, two audiences or two landing pages.
Is my data stored?
No. Everything runs in your browser.
Next stepGet in touch
Want tests that pay off?
Send a short brief. A specialist reads your account with read-only access and comes back with the three changes we would make first.
01You send a short brief about the business.
02We read the account with read-only access.
03You get three changes in writing, yours to keep either way.