> For the complete documentation index, see [llms.txt](https://docs.messageflow.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.messageflow.com/email/e-mail-campaigns/campaigns/creating-e-mail-campaigns/a-b-campaigns.md).

# A/B campaigns

*A/B campaigns: Effectiveness optimization*

A/B testing serves as an effective tool for communication optimization. It allows verifying which message variant resonates best with recipients, resulting in higher open rates (OR) and click-through rates (CTR).

The process begins by selecting the **Create campaign** button on the campaign list, providing a name, and then clicking **Create A/B campaign**. The system guides the user through a dedicated five-step workflow.

<figure><img src="/files/dCD7z1mefa7pn2X9lh75" alt=""><figcaption></figcaption></figure>

{% stepper %}
{% step %}

### Step 1: A/B settings

This section defines the foundation of the test:

* **Element to test**: The parameter to be compared. Available options include Subject, Sender, and Email creation. The Subject option verifies which title encourages more opens. The Sender option tests different sender names. The Email creation option compares different layouts, content, or CTA buttons.
* **Number of variants**: The selection ranges from 2 to 5 versions.
* **Percentage of recipients**: This setting determines the portion of the audience participating in the test, ranging from 10% to 90%. The remaining database segment automatically receives the winning version.
* **A/B Test Winner Criteria**: The system identifies the winner based on either the highest number of opens or clicks.
* **Test duration**: A specified number of hours after which the system concludes the test and distributes the winning version to the remaining recipients.

<figure><img src="/files/pBrAU2tB29aWx3PQnMzd" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### Step 2: E-mail creation

The configuration in this step depends on the selected test element:

* When testing the **Subject** or **Sender**, appropriate fields are configured for each variant. Options include personalization features or an AI assistant for title generation.

<figure><img src="/files/Xb65XynSWSguoEQM17fe" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/CjUq6cTpaHDYnuoBrL1Y" alt=""><figcaption></figcaption></figure>

* When testing **Email creation**, separate graphic designs are prepared for each version. This configuration is completed using the drag and drop editor, custom HTML code, or existing templates.

<figure><img src="/files/zDj73sJB84COgdUZ9AVx" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### Step 3: Basic information

This step involves providing information common to all variants, excluding the tested element. It includes configuring click tracking, Google Analytics integration with UTM parameters, and adding potential PDF attachments.
{% endstep %}

{% step %}

### Step 4: Recipients

The target group for the test is defined in this section. Available options include static contact lists or dynamic segments. The MessageFlow Segments feature operates as an additionally paid module available in selected packages. The system automatically divides the chosen group according to the percentages established during the first step.
{% endstep %}

{% step %}

### Step 5: Summary

This stage provides the final verification before launching the test:

* **Distribution visualization:** A pie chart displays the percentage distribution of campaign recipients between the test variants and the group designated to receive the winning version.
* **Data verification:** The summary section allows reviewing the configured subjects, senders, and the recipient count.
* **Send test:** Test messages of all variants can be dispatched to a specified address to ensure personalization and links function correctly.

<figure><img src="/files/cqPrYX5wOMsYJFdr39Os" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### A/B campaign execution

Once the configuration is complete, two final actions are available. Selecting the **Save** button preserves the campaign in draft mode for future modifications. Selecting the **Save and send** button finalizes the setup. The test then initiates immediately or at a scheduled time. From this point, the process becomes fully automated. The system dispatches the test variants, waits the specified number of hours, analyzes the results, and delivers the best-performing version to the remaining audience.
{% endstep %}
{% endstepper %}