> ## Documentation Index
> Fetch the complete documentation index at: https://help.experro.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Details

Details is the first step of the A/B test flow and the only one you must complete before the experiment exists. Saving this step creates the draft.

## Configuration Details

<Frame>
  <img src="https://mintcdn.com/experro/vL9u48aLzyWUbwlI/images/experiments/a_b_testing/imagesource_details.png?fit=max&auto=format&n=vL9u48aLzyWUbwlI&q=85&s=4039ee55a0861a0110ab330b9253752d" alt="" width="1440" height="900" data-path="images/experiments/a_b_testing/imagesource_details.png" />
</Frame>

| Field               | Required | Description                                                       |
| ------------------- | -------- | ----------------------------------------------------------------- |
| **Experiment Name** | Yes      | A name for the experiment. Names do not have to be unique.        |
| **Observation**     | No       | What you noticed that prompted the test. Up to 10,000 characters. |
| **Hypothesis**      | No       | What you expect the change to do. Up to 10,000 characters.        |
| **Channel**         | Yes      | The channel the experiment runs on.                               |
| **Language**        | Yes      | The language the experiment runs in.                              |

## Why Observation and Hypothesis Matter

These two fields are optional, but they turn a test into a record others can learn from. Six months later, the numbers alone rarely explain why anyone ran the test.

Write the observation as something you measured:

> Many users add products to the cart but do not complete the checkout process.

Write the hypothesis as a prediction you can falsify:

> Simplifying the checkout steps will reduce friction and increase completed purchases.

## Key Considerations

* **Channel** and **Language** decide which pages, components, algorithms, and rules appear later on the **Experience** step. Choose them carefully.
* Once you save this step, both fields become read-only. To change them, clone the experiment and start again.
* If the workspace has only one channel and one language, both fields are hidden and applied automatically.

<Note>
  You cannot move past Details until the name, channel, and language are set.
</Note>

## What's Next

* [Metrics](/experiments/a_b_testing/metrics)