Optimize for revenue per visitor, not just conversion rate
Omniconvert Explore scores every experiment on revenue per visitor, so you ship what actually grows revenue.
Placeholder values below. Replace them with your own control numbers and test result.
See how Explore scores every test on revenue ↓
Revenue per visitor = conversion rate × average order value. Arithmetic on the numbers you enter. It is a scenario, not a benchmark or a forecast.
Built on data from 2,500+ eCommerce brands
In shortStop optimizing conversion rate in isolation. Omniconvert Explore measures every test on revenue per visitor, so a win on conversion rate never costs you revenue.
How often could a win be costing you revenue?
Tick what is true today. Be honest, nobody is watching.
A test can win on conversion rate and still lose revenue
Revenue per visitor is conversion rate multiplied by average order value. When one goes up and the other goes down, the product can fall, even though the test looks like a win.
Illustrative example, not a benchmark. Bar lengths are not to scale across rows.
How Explore protects revenue per visitor
Three things change once every test in Omniconvert Explore is read on revenue, not just conversion rate.
Explore scores every test on revenue per visitor
- Revenue per visitor sits beside conversion rate and AOV on every variant
- The number you decide on is the number that pays the bills
- Conversion rate shows whether someone bought, revenue per visitor also shows how much
Your team still: picks the winner, now with the revenue number in front of you.
Explore catches AOV-negative wins before they ship
- Pushing cheaper products or heavier discounts can lift conversion and lower order value
- Explore combines both into revenue per visitor
- That kind of win shows up as a revenue loss, not a green arrow you act on by mistake
Your team still: decides what to test next, instead of cleaning up a shipped loss.
Explore segments revenue per visitor by traffic source
- Paid, organic, email and returning traffic each convert at a different value
- A blended number can hide a change that helps one source and hurts another
- Keep a variant where it wins and drop it where it does not
Your team still: decides which traffic matters most.
Your testing decisions, conversion rate only vs Explore
| Task | Conversion rate only | With Explore |
|---|---|---|
| Deciding metric | Conversion rate | Revenue per visitor, with conversion rate and AOV as context |
| A cheaper-product win | Looks like a win, because conversion rate is up | Shows as a revenue loss on the results screen |
| Conversion and AOV | Tracked apart, can pull opposite ways | Combined into one number on every variant |
| Traffic mix | One blended number, read as a whole | Broken down by traffic source and segment |
| What ships | Whatever lifts conversion rate | The version that grows revenue |
WatchShop decided on revenue per visitor, not conversion rate alone
WatchShop measured a product and checkout test on revenue per visitor, not conversion rate alone, and shipped the version that grew revenue across the board.
Omniconvert, WatchShop case study
Two things worth knowing upfront
Isn't a conversion rate win always a good thing?
Not if it comes from a lower-margin segment or a lower average order value. A test can lift conversion by nudging shoppers toward a cheaper product and still leave you with less revenue per visitor than the version you replaced.
Why not just track average order value separately?
Because a single test can move conversion rate and average order value in opposite directions. Revenue per visitor is the one number that combines both, so it catches a net loss that either metric on its own would hide.
Common questions
What is revenue per visitor and how is it different from conversion rate?
Revenue per visitor is defined as total revenue divided by total visitors over the same period, which is the same as conversion rate multiplied by average order value. Conversion rate only measures whether a visitor bought; revenue per visitor also measures how much they spent, so it reflects the total value each visit generates rather than just the count of orders.
How do you calculate revenue per visitor?
Divide total revenue in a period by the total number of visitors in that same period. Equivalently, multiply your conversion rate by your average order value. Both give the average revenue each visit is worth.
Why can a test win on conversion rate but lose on revenue?
A change can increase the number of people who buy while lowering how much each of them spends, for example by promoting a cheaper product or a bigger discount. Conversion rate goes up, average order value goes down, and revenue per visitor can end up flat or negative even though the test looks like a win.
What is a good revenue per visitor for eCommerce?
There is no single benchmark, because revenue per visitor depends on your average order value, margin and traffic mix. The useful comparison is against your own baseline and between test variants, not against an industry average that assumes a different catalog and price point.
Should I optimize for conversion rate or revenue per visitor?
Use revenue per visitor as the decision metric and keep conversion rate and average order value as supporting context. Conversion rate on its own can send you toward changes that add orders but subtract revenue; revenue per visitor keeps both effects in a single number.
How does Omniconvert Explore measure revenue per visitor?
Omniconvert Explore records revenue per visitor for each test variant alongside conversion rate and average order value, and lets you break every result down by traffic source and customer segment, so you can see the revenue impact of a change rather than inferring it from conversion rate alone.
See what your tests are doing to revenue
30 minutes. We'll look at your own test results on revenue per visitor, not a demo store.