> ## 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.

# Target Audience

Target Audience controls entry to the experiment. It answers two separate questions: who is eligible, and how many of the eligible actually take part.

## Targeting

Choose one of two options.

**All Visitors** makes every visitor eligible. This is the default and the right choice for most tests.

**Condition Based Target** narrows eligibility using audience segments, UTM parameters, locations, or device types. See [Condition Based Targeting](/experiments/targeting/condition_based_targeting).

## Traffic Allocation

Traffic allocation sets the percentage of eligible visitors who enter the experiment, from 1 to 100. It defaults to 100.

Lower it when you want to limit exposure — a risky change might start at 10% while you watch for problems.

<Frame>
  <img src="https://mintcdn.com/experro/vL9u48aLzyWUbwlI/images/experiments/a_b_testing/imagesource_targetaudience.png?fit=max&auto=format&n=vL9u48aLzyWUbwlI&q=85&s=fad648520f1c238afdb113fe059d8155" alt="" width="1440" height="900" data-path="images/experiments/a_b_testing/imagesource_targetaudience.png" />
</Frame>

## How Targeting and Allocation Combine

The two settings apply in order. Targeting runs first and produces the eligible pool. Traffic allocation then takes a percentage of that pool.

Suppose 1,000 visitors arrive:

1. Targeting restricts the experiment to US visitors, leaving 500 eligible.
2. Traffic allocation is set to 50%, so 250 visitors enter the experiment.
3. Those 250 are then split across the control group and variants on the **Variants** step.

The remaining 750 visitors never enter the experiment and see your live experience.

<Note>
  Traffic allocation on this step controls entry to the experiment. The split between the control group and each variant is set separately on the **Variants** step.
</Note>

## Key Considerations

* Values of zero or below are corrected to 1.
* Narrow targeting reduces the traffic reaching your test, which lengthens the time needed for a reliable result.

## What's Next

* [Condition Based Targeting](/experiments/targeting/condition_based_targeting)
* [Duration](/experiments/a_b_testing/duration)