> For the complete documentation index, see [llms.txt](https://daisychain.gitbook.io/help/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://daisychain.gitbook.io/help/texting/campaigns/a-b-testing.md).
# A/B Testing
A/B testing helps you figure out which message resonates best with your supporters. In Daisychain, you can create multiple message variants, send them to a portion of your audience, see how each variant performed, then send the winning message to the rest.
### Starting an A/B Test
1. **Create a Campaign**: Start as you normally would.
2. **Click "Create A/B Test"** button on the top-right of the Initial Message box**:**
This will then be able to access tabs in your Initial Message box where you can create different variants. By default, these tabs will be labeled "A", "B", "C", etc. You can create up to five variants.
3. **Click the A/B test settings** button to launch a pop-up window where you can:
1. Name each variant
2. Use the **Experiment Audience Percentage slider** to choose what portion of your audience will receive the test messages. For example, if you set the slider to 50%, and you have 100,000 people, then about 25,000 people will receive each of Variant A and Variant B, and the remaining 50,000 people will be held back.
4. **Review Message Variants:** on Step 3 of the Campaign Wizard you'll have a chance to ensure message previews, personalization, and counts all look right.
5. **View the Report:** After sending, you'll land on the [Campaign Report](/help/texting/campaigns/campaign-report.md), where you'll see:
* A funnel chart showing performance by variant.
* A results table with side-by-side stats for each variant:
* Delivered
* Clicks (if there’s a link)
* Conversions (if you used integrations like ActBlue, Mobilize, etc.)
* Replies
* Opt-outs
Daisychain also provides a statistical analysis indicator, which activates once each variant has at least 30 recipients.
5. **Pick a Winner:** When you’re ready, click Mark Winner next to your chosen variant. The winning message will then be sent to the remaining audience (the percentage you held back).
{% hint style="success" %}
#### Tips for A/B Testing
* **Test one element at a time:** tone, call to action, framing, use of images, etc.
* **Give the test enough time** to generate useful data (replies, clicks, etc.) before selecting a winner.
* **Apply learnings beyond the test:** use insights to improve all future outreach.
{% endhint %}
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://daisychain.gitbook.io/help/texting/campaigns/a-b-testing.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.