Conversion Rate vs Revenue Per Visitor: The Math That Matters

Conversion rate vs revenue per visitor is the fight that decides whether a winning A/B test actually makes you money. Most stores watch conversion rate because it is the easiest number on the dashboard. But conversion rate only tells you how many people bought. It says nothing about what they bought, and a test that lifts conversion rate while quietly shrinking average order value can leave you with more orders and less revenue.

Conversion rate vs revenue per visitor

Conversion rate is orders divided by visitors. Average order value is revenue divided by orders. Revenue per visitor is the two multiplied together, revenue divided by visitors. It is the only one of the three that tells you what a visitor is actually worth to the business. A page can raise conversion rate and lower revenue per visitor at the same time, if the extra orders come in at a lower price. It can also lower conversion rate and raise revenue per visitor, if it sells fewer people but sells them a better order. Revenue per visitor is the number that nets both effects out, which is why it is the number we test on.

The math that catches people out

You do not need a big sample to see the gap. PIP Wellness tested a listicle style page that funneled into their product page, against their existing control, on Meta traffic. Revenue per visitor moved from $0.40 to $0.66, a 64% lift. The dollar figures show the point plainly. Two pages can send the same 1,000 visitors, and one produces $400 while the other produces $660, without either number ever showing up as a conversion rate on a dashboard. If you only tracked conversion rate on that test, you could pick the version that converts more people into smaller orders and call it a win, while leaving $260 on the table for every thousand visitors.

What happens when you optimize for the wrong number

Conversion rate is easy to move and easy to game. Add a discount, tighten the offer, remove a step, and more people will check out. None of that tells you whether the business made more money. A discount that raises conversion rate by pulling in price sensitive buyers can lower average order value enough to erase the gain, and a store that only watches conversion rate will never see it happen, because the dashboard is reporting the wrong metric for the question being asked. Revenue per visitor cannot be gamed the same way. If a change moves people toward a cheaper option, revenue per visitor will show it, because average order value is baked into the number. That is also why we never read results off the ad platform. Ad platforms report on their own attribution window and their own definition of a conversion. We read every result off the A/B testing platform running against the brand’s own live traffic, split test, so the number reflects what actually happened on the site.

What the measured results actually show

Across our page rebuilds, the lift almost always shows up as conversion rate and order value moving together, not one canceling the other out. Sonic Brush saw a 24% revenue per visitor lift on our product page against theirs, on Google Ads traffic. Free the Roots saw 33% against their collection page, on Meta traffic. Legendary Car Care saw 32% against their product page, on Meta traffic. Bareline, a streetwear denim brand, ran our product page template against theirs across 92,484 sessions and won by 13.5%, at 99% probability to win. Gum of Gods ran our offer page against their own offer page and the gap was 97%. Every one of these is the new page against the brand’s current page, split on their own traffic, read straight from the testing platform. You can see the full set of results on /results.

Why we guarantee revenue per visitor, not conversion rate

Our guarantee is a 10% lift in revenue per visitor in a controlled test, or your money back, across up to 5 test iterations. We do not guarantee conversion rate, because conversion rate alone is not a business result. It is possible to hit a conversion rate target and still lose money on the average order. Revenue per visitor forces both halves of the equation, how many people buy and what they buy, to move in the store’s favor at the same time. That is the number we build the $2,500 page rebuild against, and it is the number worth watching on your own store before you trust any test result. The full guarantee terms are on /guarantee.