# How to create an A/B test campaign

> Compare subject lines, content, sender names or whole emails with Kelune CRM Pro, send to a sample, and let opens or clicks pick the winning version.

Product: Kelune CRM Self-hosted (hosted by you). Page: https://kelunecrm.com/docs/wordpress/email-campaigns/ab-test-campaign

An A/B test sends different versions of a campaign to a test audience, measures the response, and identifies a winning version. A/B testing requires **Kelune CRM Pro**.

> **Note:**
>
> **Plan availability:** Free users can see the A/B Testing entry point and its upgrade prompt. An active Kelune CRM Pro license unlocks the test configuration and reporting workflow documented below.

Create and save the campaign first, then open its A/B test configuration.

## Configure the test

1. Enable **A/B Testing**.
2. Choose the **Winner Selection Metric**: **Open Rate** or **Click Rate**.
3. Set the **Test Sample Size** between 10% and 100%.
4. Set the **Winner Wait** period in hours.
5. Add one or more test variants.

![A/B Testing Configuration drawer with sample size and variants](https://kelunecrm.com/images/docs/wordpress/email-campaigns/ab-test-campaign/ab-test-config.webp)

Choose **Open Rate** when you are primarily testing the subject line or sender name. Choose **Click Rate** when the meaningful difference is in the email content or call to action.

## Create each variant

Give the variant a short label, then choose its **Test Type**:

- **Subject Line** changes the subject only.
- **Email Content** changes the body of the message.
- **From Name** changes the displayed sender name.
- **Full Email** changes the subject and content.

Set the variant's share of the test sample. The original campaign receives the remaining share, so check that the allocation produces the comparison you intend.

> **Tip:**
>
> Change one meaningful element at a time when you want a clear explanation for the result. A **Full Email** test can identify a winner, but it may not tell you which individual change caused the improvement.

## Understand the sample and winner

The test sample receives the original and variant messages according to their assigned shares. After the winner wait period, Kelune CRM compares the selected metric and can send the winning version to the remainder of the campaign audience.

If the test sample is 100%, the entire audience is used for the comparison and no remainder is left to receive the winner. Use a smaller sample when the purpose is to optimize delivery to the rest of the audience.

## Start and monitor the test

Preview and test-send every version before activating the campaign. Confirm that the subject, sender, links, content, and sample allocation are correct.

Once sending starts, the test configuration becomes read-only. Use the campaign analytics to compare the original and variants. When the test state permits it, you can also use the manual winner action instead of waiting for automatic selection.

> **Caution:**
>
> Do not select a winner from a very small response unless timing makes it necessary. A difference based on only a few opens or clicks may not represent how the wider audience will behave.