Overview

This article provides an overview of experiments in batch messages, automated messages, and Podium orchestrations, highlighting some key differences.

Batch message experiments

In batch messages, you can create subject line and message body experiments to optimize engagement.

• Learn about subject line experiments in batch messages here.
• Learn about message body experiments in batch messages here.

You can view the performance of these experiments in the left sidebar of the message creation page, under Message Performance.

The ability to add a control variant at the message body level is only available in batch messages.

Automated message experiments

Automated message experiments take advantage of machine learning to automatically favor better performing variants over time using the multi-armed bandit strategy. This approach is somewhat different than the traditional A|B split test, but it accomplishes the same objective with more efficiency and less work for you. As with batch messages, you can create subject line and message body experiments.

• Learn about automated message experiments here.

You can view the performance of automated message experiments in the left sidebar of the message creation page, under Message Performance.

The multi-armed bandit strategy doesn't allow for control variants.

Podium orchestration experiments

Experiments in Podium orchestrations provide a simple yet powerful way to create and test multiple automations against one another—with the ability to select the best performing automation.

• Learn more about Podium orchestration experiments here.

In order to use a control variant in a Podium experiment, the experiment must be built out at the Podium level. Message body experiments in Podium use the multi-armed bandit strategy, which doesn't allow for control variants.