Segments

A segment is a definition of a subpopulation, like a certain cohort of users. Define it once and reuse it across multiple flags and experiments, so your targeting stays consistent and you don’t redefine the same criteria each time.

​

Segment Components

Segments consist of:

• Inclusion criteria: Filters that describe which users belong to the segment
• Allocation: The percentage of users matching the criteria to include
• Exclusivity: Optional setting to make segments mutually exclusive

These components follow the shared audience targeting model that flags and experiments also use.

​

Concepts

• Materialized Segments: Segments that load their list of units from BigQuery.
• Audience Targeting: Inclusion criteria, allocation, randomization, sticky assignment, and exclusivity for rules, experiments, and segments.

​

Tasks

• Create Segments: Create a segment to define a reusable group of users.
• Define the Audience: Add inclusion criteria to target specific users in rules and segments.
• Create Materialized Segment: Create a materialized segment that loads units from BigQuery.
• Launch an A/B Test: Target segments in experiments.