User Segment-Based Feature Targeting

Segment-Based Targeting

User segments are named groups of users that can be targeted across multiple feature flags. Instead of listing individual user IDs in each flag, you define segments once and reference them everywhere.

Configuration Example

{
  "ai-assistant": {
    "name": "AI Assistant Feature",
    "description": "GPT-powered assistant in the help center",
    "type": "boolean",
    "enabled": true,
    "defaultValue": false,
    "targeting": [
      {
        "type": "user-segment",
        "segment": "beta-testers"
      }
    ]
  }
}

Defining Segments

Segments are typically defined separately from flags. A segment has its own rules:

{
  "segments": {
    "beta-testers": {
      "name": "Beta Testers",
      "rules": [
        { "attribute": "plan", "op": "in", "values": ["enterprise"] },
        { "attribute": "opted_in_beta", "op": "equals", "values": ["true"] }
      ],
      "included": ["user-001", "user-042"],
      "excluded": ["user-099"]
    }
  }
}

Segments vs User IDs

Aspect	User IDs	Segments
Scale	1-100 users	Unlimited (rule-based)
Maintenance	Per-flag management	Central management
Reusability	Single flag only	Across all flags
Dynamic	Static list	Rule-based membership
Audit trail	Change per flag	Change in one place

Common Segment Definitions

• beta-testers: Users who opted into the beta program
• internal-team: Employees identified by email domain
• premium-users: Users on paid plans
• power-users: Users exceeding a usage threshold
• early-adopters: Users who signed up before a certain date

LaunchDarkly Segment Rules

In LaunchDarkly, segments use segmentMatch in flag rules:

{
  "rules": [
    {
      "clauses": [
        {
          "attribute": "segment",
          "op": "segmentMatch",
          "values": ["beta-testers"]
        }
      ],
      "variation": 0
    }
  ]
}

Best Practices

• Name segments descriptively (enterprise-beta not group-1)
• Keep segment definitions close to your flag management code
• Review segment membership regularly; remove inactive users
• Use rule-based segments over static user lists when possible