Skip to main content

Creating a Feature Flag

Learn how to create and configure feature flags in the Toggly dashboard.

Quick Create​

The fastest way to create a feature flag:

  1. Navigate to Features in the dashboard
  2. Click Create Feature Flag
  3. Enter a feature key (e.g., new-checkout-flow)
  4. Click Create

The feature flag will be created with default settings and disabled state.

Detailed Creation​

For more control, use the detailed creation form:

Step 1: Basic Information​

  • Feature Key: Unique identifier (required)

    • Use kebab-case: new-checkout-flow
    • Must be unique within the environment
    • Cannot be changed after creation

  • Name: Human-readable name (optional)

    • Display name in the dashboard
    • Can be changed later

  • Description: Detailed description (recommended)

    • Explain what the feature does
    • Include context and use cases
    • Helpful for team members

Step 2: Feature Type​

Choose the feature flag type:

  • Boolean: Simple on/off switch
  • Multi-Variant: Returns different values (variants)

For multi-variant flags, define variants:

  • Control: Default variant (e.g., control)
  • Variant A: First treatment (e.g., red-button)
  • Variant B: Second treatment (e.g., blue-button)
  • Variant C: Additional variants as needed

Step 3: Initial State​

Set the initial state:

  • Enabled: Feature is active (use with caution)
  • Disabled: Feature is inactive (recommended)
Start Disabled

Always start with the feature disabled. Enable it after configuring targeting rules and testing.

Step 4: Targeting Rules​

Configure who sees the feature:

  • All Users: Enable for everyone
  • Percentage: Enable for X% of users
  • Segments: Enable for specific user segments
  • Custom Rules: Complex targeting conditions

Step 5: Environment​

Select the environment:

  • Development: For local development
  • Staging: For pre-production testing
  • Production: For live users

Feature Key Best Practices​

Naming Conventions​

Use descriptive, consistent naming:

// Good examples
'new-checkout-flow'
'enhanced-search-results'
'premium-dashboard'
'beta-mobile-app'

// Bad examples
'feature-1'
'new-thing'
'flag-abc'

Organization​

Use prefixes or tags to organize:

  • frontend-*: Frontend features
  • backend-*: Backend features
  • experiment-*: Experimental features
  • critical-*: Critical features

Multi-Variant Flags​

Multi-variant flags return different values instead of just true/false:

Creating Variants​

  1. Select Multi-Variant as the feature type
  2. Click Add Variant
  3. Enter variant key (e.g., red, blue, green)
  4. Add variant description
  5. Repeat for each variant

Variant Distribution​

Configure how variants are distributed:

  • Equal Distribution: Each variant gets equal traffic
  • Weighted Distribution: Assign custom weights
  • Targeted Distribution: Assign variants to segments

Example:

// Variant distribution
control: 50%  // 50% of users see control
red: 25%      // 25% see red variant
blue: 25%     // 25% see blue variant

Targeting Configuration​

All Users​

Enable the feature for everyone:

Targeting: All Users
No additional configuration needed

Percentage Rollout​

Gradually roll out to a percentage:

Targeting: Percentage Rollout
Percentage: 25%  # Enable for 25% of users

User Segments​

Target specific segments:

Targeting: User Segments
Segments:
  - premium-users
  - beta-testers

Custom Rules​

Create complex targeting:

Targeting: Custom Rules
Rule: (plan = 'premium' AND country = 'US') OR (plan = 'enterprise')

After Creation​

Once created, you can:

  1. Test the Flag: Use the SDK to evaluate the flag
  2. Configure Targeting: Set up targeting rules
  3. Monitor Metrics: Watch usage in the dashboard
  4. Enable When Ready: Enable after testing

Example: Complete Feature Flag​

Here's an example of a fully configured feature flag:

Feature Key: new-checkout-flow
Name: New Checkout Flow
Description: Redesigned checkout experience with improved UX
Type: Boolean
State: Disabled
Environment: Production

Targeting:
  Type: Percentage Rollout
  Percentage: 10%
  Segment: premium-users (optional)

Tags:
  - frontend
  - checkout
  - experiment

Next Steps​