Iterable

Subsets supports syncing data to Iterable using the /api/users/bulkUpdate endpoint.

Setup Guide

To set up the Iterable integration:

1. Verify Custom Field Capacity

   • Confirm that your Iterable account can accommodate 5 additional custom user profile fields
   • Subsets will create these fields automatically during setup

2. Create an API Key in Iterable

   • Navigate to your Iterable account settings
   • Create a new Server-side API key with user profile update permissions
   • Follow Iterable's server-side API key documentation

3. Provide API Key to Subsets

   • You will get access to the Subsets platform where you can add the key

4. Testing

   • Provide access to a sandbox enviroment or create test users in your production enviroment.
   • This will be used to verify the integration is working correctly.

Custom User Profile Fields

Subsets creates the following custom data fields in your Iterable user profiles:

Field Name	Description	Type
subsetsExperimentId	Unique identifier for the experiment	String
subsetsExperimentAddedAt	Timestamp when user was added to experiment	String
subsetsExperimentName	Name of the experiment	String
subsetsSubscriberChurnRisk	Subsets probability subscriber will churn within 30 days.	String
subsetsExperimentAssignment	Indicates if the subscriber is in the experiments treatment or control group.	String

Note: All fields are prefixed with subsets to avoid conflicts with existing user profile fields in your Iterable account.

Data Synchronization

• Batch Size: 1,000 users per API call
• API Endpoint: https://api.iterable.com/api/users/bulkUpdate
• Update Method: Merge nested objects to preserve existing user data
• Frequency: Based on experiment updates and schedule configured with Subsets
• User Identification: Uses userId to identify users in Iterable

Implementation Notes

• Fields are automatically created by Subsets when the integration is first configured
• Updates include experiment assignments, churn risk scores, and audience information
• The integration handles batch processing to efficiently update large user sets
• All timestamps are stored in UTC format