Skip to content
Getting started

A/B testing campaigns

Stop guessing which subject line or design wins — test it. An A/B campaign splits your recipients across variants, measures how each performs, and lets the data decide.

5 min read

What you'll need

  • A campaign audience (list or segment) large enough to split meaningfully
  • Two or more ideas to compare — subjects, templates or both

How an A/B campaign works

You create a campaign as an A/B test and add variants. Each variant carries a weight that determines its share of the audience, and each contact is assigned to exactly one variant — the same way every time, so results stay stable and un-skewed.

  • Variants — two or more versions to compare.
  • Weight — each variant's share of the split (e.g. 50/50, or a small test slice per variant).
  • Stable assignment — a contact always lands in the same variant, so a resend won't scramble your test.

What you can vary

  • Subject line — the most common test, and often the biggest lever on opens.
  • Template — test an entirely different design or layout per variant.
  • Or both — give each variant its own subject and template.

Pick the winner

Set which metric decides the winner — opens or clicks — and how long the test runs before that decision. When the window closes, the winning variant is the one that led on your chosen metric.

  1. 1

    Create an A/B campaign

    Start a new campaign and choose the A/B type instead of a regular send.

  2. 2

    Add your variants

    Add two or more variants, set each one's weight, and give it a subject and/or template.

  3. 3

    Choose the winning metric and window

    Pick opens or clicks as the winning metric and set how long the test should run.

Keep tests honest: change one thing at a time. If you vary both the subject and the template, you'll know which variant won — but not why.

Frequently asked questions

What should I test first?

Subject lines — they're the biggest lever on opens and the easiest to compare cleanly. Move on to content once your subjects are dialed in.

How is a contact assigned to a variant?

Deterministically by weight — the same contact always gets the same variant, so results aren't skewed by resends.

How is the winner decided?

By the metric you choose (opens or clicks), measured over the test window you set.

Start sending in minutes

Create a free account, connect your domain and reach your audience across every channel — no credit card needed.

Free plan forever · No credit card required · Set up in minutes

A/B test email campaigns: variants, split & winner metric — Climails