Revenue per Visitor Calculator

How the figure is built

Revenue per visitor is revenue divided by sessions: revenue ÷ sessions. Equivalently it is conversion rate × average order value, which is why it is the better single metric than either.

It collapses the two things a site can influence into one number, so a change that trades one for the other shows its net effect immediately.

The contribution version, contribution per visitor. Is what belongs in any comparison against traffic cost.

A real example

$58,000 from 40,000 sessions is $1.45 per visitor. In contribution terms, $31,900 ÷ 40,000 = $0.80.

That $0.80 is the maximum that can profitably be paid for a session, which converts directly into a maximum cost per click.

Now a test that raises conversion from 2.5% to 2.9% but lowers order value from $58 to $49 through a discount: revenue per visitor becomes $1.42, slightly worse despite the conversion improvement.

The conversion metric alone would have declared that test a winner. Revenue per visitor catches it, which is why it is the figure to run tests against.

The usual mistakes

It averages across traffic sources with very different value, so a shift in channel mix moves it without any change in site performance.

It also ignores lifetime value entirely. A visitor who buys a small first order and becomes a repeat customer is worth more than the session suggests, and businesses judging traffic purely on immediate revenue per visitor will undervalue their best acquisition sources.

Using the result

Use it as the primary success metric for A/B tests, since it prevents the common failure of optimising conversion at the expense of basket size.

Then calculate it per channel and compare against that channel's cost per session. The two together determine which traffic is worth buying and how much.

Why it is the right number for testing

A test that changes anything about the purchase path affects both conversion and order value, and the two frequently move in opposite directions. Free shipping thresholds, bundle prompts, urgency messaging and discount offers all do this.

Measuring only conversion declares a winner whenever more people buy, regardless of what they bought. Measuring only order value declares a winner whenever baskets grow, regardless of how many were abandoned.

Revenue per visitor is the only metric that nets the two, and using it as the default test metric prevents a category of decisions that improve one number while reducing the money. It costs nothing to adopt and it changes the outcome of a meaningful share of tests.

Comparing it against cost per session by channel turns it into a direct profitability measure, since a channel where cost exceeds contribution per visitor is losing money regardless of how good its conversion rate looks.

That comparison also identifies channels worth expanding, which the conversion rate alone cannot, because a low-converting channel with cheap traffic can be the most profitable one you have.

Tracking it weekly rather than monthly catches deterioration from a broken template or a failing script within days, which is the sort of problem that otherwise runs for a month before anyone notices.