Optimizely

A/B testing and experimentation platform with a REST API for managing projects, experiments, campaigns, and results.

Capabilities

Integration	Available	Notes
API	✓	Projects, Experiments, Campaigns, Audiences, Results
MCP	-	Not available
CLI	✓	optimizely.js
SDK	✓	JavaScript, Python, Ruby, Java, Go, C#, PHP, React, Swift, Android

Authentication

• Type: Bearer Token (Personal Access Token or OAuth 2.0)
• Header: Authorization: Bearer {personal_token}
• Get token: https://app.optimizely.com/v2/profile/api > Generate New Token

Common Agent Operations

List Projects

GET https://api.optimizely.com/v2/projects

Get Project

GET https://api.optimizely.com/v2/projects/{project_id}

List Experiments

GET https://api.optimizely.com/v2/experiments?project_id={project_id}

Get Experiment

GET https://api.optimizely.com/v2/experiments/{experiment_id}

Get Experiment Results

GET https://api.optimizely.com/v2/experiments/{experiment_id}/results

Create Experiment

POST https://api.optimizely.com/v2/experiments

{
  "project_id": 12345,
  "name": "Homepage CTA Test",
  "type": "a/b",
  "variations": [
    { "name": "Control", "weight": 5000 },
    { "name": "Variation 1", "weight": 5000 }
  ],
  "metrics": [{ "event_id": 67890 }],
  "status": "not_started"
}

Update Experiment

PATCH https://api.optimizely.com/v2/experiments/{experiment_id}

{
  "status": "running"
}

List Campaigns

GET https://api.optimizely.com/v2/campaigns?project_id={project_id}

Get Campaign Results

GET https://api.optimizely.com/v2/campaigns/{campaign_id}/results

List Audiences

GET https://api.optimizely.com/v2/audiences?project_id={project_id}

List Events

GET https://api.optimizely.com/v2/events?project_id={project_id}

List Pages

GET https://api.optimizely.com/v2/pages?project_id={project_id}

Key Metrics

Experiment Results

• variation_id - Variation identifier
• variation_name - Variation display name
• visitors - Unique visitors per variation
• conversions - Conversion count
• conversion_rate - Rate as decimal
• improvement - Percentage improvement vs. control
• statistical_significance - Confidence level
• is_baseline - Whether this is the control

Experiment Properties

• name - Experiment name
• status - not_started, running, paused, archived
• type - a/b, multivariate, personalization
• traffic_allocation - Percentage of traffic (0-10000 = 0-100%)
• variations - Array of variations with weights

Parameters

List Experiments

• project_id (required) - Project to list experiments for
• page - Page number
• per_page - Results per page (default: 25)
• status - Filter by status

Get Results

• start_time - Results start time (ISO 8601)
• end_time - Results end time (ISO 8601)

Create Experiment

• project_id (required) - Parent project
• name (required) - Experiment name
• type - Experiment type (default: a/b)
• variations (required) - Array of variations with name and weight
• metrics - Array of metric/event configurations
• audience_conditions - Targeting conditions
• traffic_allocation - Traffic percentage (0-10000)

When to Use

• Running A/B tests on web pages and features
• Managing experimentation programs at scale
• Pulling experiment results for analysis
• Automating experiment creation and monitoring
• Feature flag management
• Personalization campaigns

Rate Limits

• 50 requests/second per personal token
• Pagination via page and per_page parameters
• OpenAPI spec available at https://api.optimizely.com/v2/swagger.json

Relevant Skills

• ab-test-setup
• page-cro
• landing-page
• personalization
• analytics-tracking