Promi offers a built-in A/B testing platform to measure the impact and statistical significance of your campaigns. These experiments can be set up to compare personalized discounts vs. no discount, personalized discounts vs. non-personalized discounts, or non-personalized discounts vs. no discount. Once a Promi campaign is set up with a control group, data will automatically be collected and added to the performance dashboard.

Performance Dashboard

This dashboard has three main sections. The "Aggregate Performance" card shows aggregate metrics from during and after the experiment period.

The "Experiment Performance" card show metrics specifically for the duration of the experiment.

Please note that the visitor count shown for treatment and control groups may be slightly different than the total visitor count shown in Shopify's reporting. This is due to Shopify using it's own logic for removing bot traffic or bounced traffic from the visitor count. Promi will also remove some bot traffic from the number shown, but the specific approach may differ between Shopify and Promi.

The "Experiment Setup" card shows metadata on the campaign, specifically the duration of the experiment vs. overall campaign.

It is customary for merchants to run an experiment until they reach a statistically significant result. If the treatment group's metrics are favorable to the control group, the merchant will shut off the experiment and keep the treatment group running to 100% of traffic.