Reading Tests Results

Last updated: July 20, 2026

Last updated: July 2026

Overview

The Tests dashboard gives you a real-time view of how each variant is performing. This article explains what each metric means, how to read the confidence score, and when you have enough data to act.

Metrics Shown Per Variant

For each variant, Jurni displays:

  • Sessions — Total unique visits routed to this variant.

  • CVR (Conversion Rate) — Percentage of sessions that resulted in a purchase.

  • AOV (Average Order Value) — Average revenue per order on this variant.

  • Revenue — Total revenue attributed to this variant.

  • RPS (Revenue Per Session) — Revenue divided by sessions. This is often the most useful single number for comparing variants because it captures both CVR and AOV together.

  • Add-to-Cart Rate — Percentage of sessions that added a product to cart.

  • Checkout Rate — Percentage of sessions that reached checkout.

Statistical Significance

Statistical significance is displayed as a confidence score from 0 to 100%.

  • 95% or above — Results are statistically significant. The difference between variants is unlikely to be due to random chance. You can act on this data with confidence.

  • 80 to 94% — Trending, but not conclusive. Collect more data before declaring a winner.

  • Below 80% — Insufficient data. Do not draw conclusions yet.

Time to reach significance depends on your traffic volume. High-traffic funnels may hit 95% within a few days. Low-traffic funnels may need two to four weeks. Running your experiment for at least seven days is recommended regardless of traffic, to account for day-of-week variations in shopper behavior.

Tip: RPS is often a better decision metric than CVR alone. A variant with slightly lower CVR but higher AOV can still outperform on total revenue.

What Results Are Not

  • Results are not retroactively adjusted. Data recorded before an allocation change is preserved as-is.

  • Results for external URL variants may not include revenue if your external tracking is not configured. See Running Experiments with External Destinations.

Frequently Asked Questions

How often do metrics update?

Metrics update in near-real-time as sessions and purchase events come in.

My CVR looks lower than usual. Is something wrong with the test?

Not necessarily. Check that both funnels are published and that the Smart Link is correctly set as your ad destination. You can also cross-reference session counts with your ad platform's click data.

Can I export results?

Export is not currently available directly from the Experiments dashboard. You can manually copy metrics for reporting purposes.

What if both variants have the same CVR?

Compare RPS and AOV. If both are truly equal after reaching statistical significance, the experiment is inconclusive. You may want to run a new test with a more significant change.