Disclaimer: This website requires Please enable JavaScript in your browser settings for the best experience.

The availability of features may depend on your plan type. Contact your Customer Success Manager if you have any questions.
Dev guideRecipesAPI ReferenceChangelog
Dev guideAPI ReferenceRecipesChangelogUser GuideGitHubDev CommunityOptimizely AcademySubmit a ticketLog In
API Reference

Fetch experiments associated with a holdout

Retrieves a list of experiments that are connected to a specific holdout through the system's data relationships.

What you'll get:

  • Experiment identification: Layer experiment IDs from connected rules
  • Timing information: Start and end times from experiment reports
  • Variation details: Deployed variation IDs when configured
  • Clean data: Filtered to remove empty or null values for clarity

How experiments connect to holdouts: This endpoint traverses complex database relationships to find experiments associated with a holdout:

  1. Holdout → Project: Matches by project_id
  2. Project → Flags: All flags in the same project
  3. Flags → Rulesets → Rules: Rules within flag configurations
  4. Rules → Reports: Experiment reports linked to rules
  5. Rules → Variations: Deployed variations (when configured)

Data relationships explained:

  • Each returned experiment represents a rule that's potentially affected by the holdout's traffic allocation
  • Experiment timing comes from associated reports (start_time, end_time)
  • Deployed variations are included when rules have configured variations
  • Results are automatically filtered to exclude empty/null values

Use cases:

  • Impact analysis: Understanding which experiments are affected by a holdout
  • Experiment coordination: Viewing overlapping experiment timelines
  • Traffic allocation planning: Seeing experiment relationships before making holdout changes
  • Reporting integration: Connecting holdout data with experiment performance
  • Debugging: Investigating unexpected experiment behavior related to holdouts

Performance considerations:

  • Query involves multiple table joins across the experiment system
  • Results are processed to remove empty values for cleaner responses
  • Response size depends on the number of connected experiments

This endpoint provides crucial visibility into the complex relationships between holdouts and the experiments they may influence.

Language
Credentials
Click Try It! to start a request and see the response here!