Define your experiment's audience

After you define the events that make up your experiment, you can set who's eligible for the experiment. You can choose to open eligibility up to all users or you can target specific groups of users.

Targeting groups of users limits experiment exposure to users in geographical locations, specific demographic groups, or who meet certain usage thresholds in your product (like power users). Defining your audience is also known as targeting segments. The segments are the unique characteristics of your audience that you want to target, such as geolocation or device type. All Amplitude user properties and cohorts are available to use in defining user segments. The value for any user property included in a rule-based user segment is always the most recent value received by Amplitude. For more information on segments, go to Segment Overview.

There is no limit on the number of user segments you can include here. Any user who belongs to more than one segment included in an experiment is assigned to the first one they match. 

Defining your audience isn't just about the segments or user properties. You must also specify your Distribution and Rollout.

• Distribution: The percentage between the control and the experiment group. By default, this is set to 50% distribution so that half of your group is the control and half is the experiment. You can customize the percentage if you want more people in one distribution group or the other. This is the distribution across your targeted audience and not your total user base.
• Rollout: The percentage of people, within your targeted audience who you want to receive the experiment. If you set this number to 100%, then all users within your targeted audience receives the experiment as either the control or treatment group. Experiment displays the potential number of people who will receive the experiment.

To define the audience for your experiment

1. Either create a new experiment or open an existing one.
2. In the Targeting section, click the edit icon.
3. In the Audience tab, select the bucketing unit you want. Typically, the bucketing unit is by User. However, you can also bucket by groups.
4. Select who you want to be eligible for this experiment. You can select All Users or Targeted Users. Targeting all users means that every users who triggers the exposure event receives your experiment. Targeting users lets you specify segments of users to receive your experiment.
5. Select the bucketing you want to target. By default, this is Amplitude ID. However you could select something like:
   • Device type
   • Language
   • Region
   • Postal code
6. Click the Distribution tab.
7. If you want to customize the distribution percentage, click Customize and then specify the percentages you want.
8. Click the Rollout tab.
9. In the All users field, specify the percentage of users you want to receive the experiment.
10. Click either Save & Close or Testers.

After you have specified your audience, specify your testers.

Testers

Testing your experiment is important so that you can be sure that all aspects of your experiment are working and to ensure that the implementation of your experiment is successful. Go to Test and Launch Your Experiment for more details about testers.

Boolean logic in segments

Amplitude applies Boolean AND logic to conditions within the same segment. These conditions are evaluated as if, else if. For example, you have two Segments:

Segment 1: Users in India and who access your site through the Web. Segment 2: Users who use Android devices and access your website through the mobile browser.

Using Boolean logic, both conditions for a segment must be met to include users in the experiment. If users access your website but aren't located in India, they're not included in Segment 1. Likewise, if users access your site through an iOS device, they aren't included in Segment 2.

The if, else if evaluation means that a user is evaluated if they meet the conditions for Segment 1. If not, they're evaluated if they meet the conditions for Segment 2.