A/B Split

This powerful tool is designed to optimize your email, push, and on-site messaging by testing two variants of content in a single one-time campaign. Instead of continuously redirecting traffic, A/B Split works by sending each variant to a small portion of a predefined audience, for best results this audience should be large enough (e.g., at least 100,000 recipients). After the initial send, the system waits for a set evaluation period to measure engagement and conversion rates. Once the winning variant is determined, the remaining audience automatically receives the more effective content. Whether you're fine-tuning emails, personalizing push notifications, or enhancing on-site interactions, A/B Split enables data-driven decisions to boost the success of your one-time campaigns.

Configuration begins by choosing the appropriate channel. Depending on the channel selected, the configuration settings may vary slightly to accommodate specific channel requirements.

A/B Split - Channel

You’re preparing an email campaign to promote your new product launch, but you're not sure which subject line will drive more opens. To find out, you use the A/B Split node in your campaign flow. You set Variant A with the subject line “Introducing Our New Collection – Just Launched!”, and Variant B with “You Asked, We Delivered – Shop the New Arrivals!”.

You configure the test by sending each version to 5% of your audience (5% for Variant A, 5% for Variant B). You then set the remaining 90% to be sent to the winning variant based on open rates. In the optimization section, you enter 120 minutes as the test duration, meaning the system will track which subject line performs better over a two-hour period. After the test ends, the email with the higher open rate will automatically be sent to the remaining 90% of your audience.

This way, your campaign is optimized in real time, and most of your users receive the version that’s proven to work better.

Distribution Allocation

• At the top, you'll see green 🟩 and blue 🟦 bars indicating the percentage of your audience that will receive each variant of the contents (Variant A and Variant B). Adjust the percentages as needed, keeping in mind that the remaining percentage (as indicated by "90% goes to winner") will be allocated to the variant that performs the best.

Selecting Content and Setting Distribution for Each Variant

• Identify Variant A 🟩 and Variant B 🟦 , use Select Content button to assign a content for each variant.
• The percentage field allows you to specify what portion of your test audience will receive this variant.
  Use the Change Content button if you wish to select a different content for this variant.

A/B Test Optimization
In this section, you define how the winning variant will be determined.

• Enter the time (e.g., 120 minutes) in the provided field, after which the system will evaluate the performance and automatically send the winning variant to the rest of the audience.