Conversion rate is the share of visitors, sessions or users who complete a goal you define, such as a purchase, signup or form submission, expressed as a percentage. The formula is conversions divided by the number of visitors (or sessions), multiplied by 100. A store with 327 orders from 25,935 visitors has a conversion rate of about 1.26%.
Conversion rate definition
The formula is simple: conversion rate = conversions / visitors x 100. The hard part is agreeing on what goes on each side of the division.
The top number is whatever action counts as a conversion for that page: an order, a trial start, a demo request, a newsletter signup.
The bottom number is the group that had the chance to convert, and tools count it differently.
| Denominator | What it counts | Typical effect |
|---|---|---|
| Sessions | Each visit separately | Lower rate, because one buyer may visit several times before buying |
| Users or visitors | Each person (or browser) once | Higher rate for the same orders |
| Test participants | Visitors bucketed into an A/B test | Only comparable within that test |
Google Analytics 4 reports both versions. Google's metric reference defines session key event rate as sessions with a key event divided by total sessions, and user key event rate as users who performed a key event divided by total users.
Always say which one you mean before comparing numbers.
Conversion rate meaning in CRO and A/B testing
In conversion rate optimization, conversion rate is usually the primary metric of a test.
Visitors are split between a control and a variant, and the test compares each group's rate.
The difference is reported as lift, which is almost always relative.
Going from 1.135% to 1.366% is a 0.231 percentage-point change but a 20.4% relative lift.
Confusing the two is one of the most common reporting mistakes.
Low baselines also make tests slow. A page converting at 0.4% needs far more traffic to detect the same relative change than one at 5%, which is why knowing your baseline conversion rate comes before planning any test.
Conversion rate on its own can mislead. A change can raise the share of people who buy while lowering what each buyer spends, so most ecommerce tests also track revenue per visitor and average order value.
Examples from real A/B tests
Three tests from our database, all run by the agency Convertibles, show how conversion rate is reported in practice:
- Sticky mobile CTA on a subscription homepage: conversion went from 1.135% to 1.366% (+20.4%) across 25,935 visitors and 327 orders, with a 95.1% probability to beat control. Revenue per visitor rose 19.6%, though that metric reached only 87.3% probability.
- Star rating under the price plus member reviews: conversion went from 4.725% to 5.460% (+15.56%) across 15,615 visitors and 795 orders, with a 98% probability of being best. Average order value stayed flat, so revenue per visitor moved with it (+15.79%).
- Cart drawer instead of an add-to-cart banner: conversion went from 0.382% to 0.444% (+16%) across 84,005 visitors and 347 orders, at 86-92% probability to beat control. On mobile, conversion rose 26% but average order value fell 21%, leaving revenue per visitor flat. That shows why conversion rate needs a revenue check beside it.
ConvertiblesConvertiblesConvertiblesNote the range of baselines: 0.38% for a high-ticket equipment store and 4.7% on a subscription product page.
A "good" conversion rate depends on price, traffic source and page type, so compare against your own history first.
You can browse more results by metric in the full test database.
Related terms
- Baseline conversion rate: the rate before a change, used to size tests.
- Lift: the relative change in conversion rate between variant and control.
For ways to move the number, read the deeper guide on how to increase conversion rate, or see every term in the CRO glossary.