Revenue per visitor calculator

Measure revenue impact at the visitor level, not only among people who purchased.

Use visitor-level revenue for launch decisions

Revenue per visitor captures conversion-rate and order-value movement in one metric tied to exposed traffic.

1. Exposure

• Keep every visitor represented: RPV includes visitors who did not purchase, so it preserves the experiment denominator.

2. Variance

• Use real revenue spread: Revenue data is skewed. Uploading representative baseline data makes planning more realistic.

3. Decision

• Compare against business value: Choose an MDE that reflects the smallest visitor-level revenue movement worth shipping.

Use this focused view when revenue per exposed visitor is the primary business outcome.

What RPV tells you

RPV is often the better ecommerce launch metric because it combines buying rate and basket size.

• Mean revenue per visitor: Average revenue across all exposed visitors in each variant.
• Confidence interval: The plausible range for the visitor-level revenue difference after accounting for observed variance.
• SRM check: Traffic allocation remains a prerequisite for trusting revenue results.

Revenue tests are variance-sensitive

A few large orders can move the mean.

• Use representative revenue data when planning.
• Check outliers and instrumentation before interpreting a large revenue swing.
• Use conversion rate and AOV as supporting diagnostics when RPV moves.

RPV methodology

The calculator treats revenue per visitor as a continuous metric and uses observed variance where visitor-level data is available.

• Continuous analysis: Revenue outcomes use continuous-metric analysis rather than a binary conversion-rate model. Use this when each visitor has a revenue value, including zero for non-buyers.
• Sample planning: Required sample depends on baseline RPV, target MDE, confidence, power, and revenue variance. Use this before launch to decide whether a revenue test is feasible.
• Guardrail context: Conversion rate and AOV can explain why RPV changed, but RPV keeps the business impact tied to exposure. Use this when the launch decision depends on total revenue impact.

Inputs for visitor-level revenue tests

Use these definitions consistently across control and variants.

• Revenue per visitor: Total revenue divided by exposed visitors. Non-buyers contribute zero revenue.
• Revenue MDE: The smallest visitor-level revenue change worth detecting.
• Revenue data: Visitor-level revenue observations used to estimate variance for planning and analysis.

Related calculators

• Average order value calculator: Evaluate order-size changes without losing the visitor-level view needed for experimentation.
• Conversion rate calculator: Plan and analyze conversion-focused A/B tests with the calculator set to binary conversion outcomes.
• A/B test sample size calculator: Find out how many visitors you need before your results are worth acting on.
• A/B test duration calculator: Find out how long to run your test before the results are worth trusting.

Common questions about RPV tests

RPV keeps ecommerce test decisions tied to exposed traffic.

• What is revenue per visitor? Revenue per visitor is total revenue divided by the visitors exposed to each variant.
• Why use RPV instead of AOV? RPV includes non-buyers, so it captures both conversion-rate and order-value changes.
• Why do RPV tests often need larger samples? Revenue is more variable than binary conversion data, and higher variance increases the sample needed.
• Can I still look at AOV? Yes. AOV is useful diagnostic context, but RPV should usually remain the primary revenue decision metric.