Skip to main content

Grouping Engagement

Learn how to group Engagement by user attributes for personalized experiences.

Overview​

Grouping allows you to create separate state machines for different user segments, enabling personalized Engagement flows.

Basic Grouping​

Group by a single attribute:

{
  "groupByConfig": {
    "groupByKeys": ["userId"]
  }
}

This creates separate state machines for each unique userId.

Multiple Attributes​

Group by multiple attributes:

{
  "groupByConfig": {
    "groupByKeys": ["userId", "userType"]
  }
}

This creates separate state machines for each unique combination of userId and userType.

Limiting State Machines​

Limit the number of active state machines:

{
  "groupByConfig": {
    "groupByKeys": ["userId"],
    "maxActiveStateMachineCount": 100
  }
}

When the limit is reached, new state machines won't be created until old ones are reset.

Context Parameters​

Store group by keys in state machine context. See State Machine DSL Overview for detailed information about context parameters.

Example: User-Specific Welcome​

{
  "ctaId": "personalized-welcome",
  "rule": {
    "groupByConfig": {
      "groupByKeys": ["userId"],
      "maxActiveStateMachineCount": 1000
    },
    "contextParams": ["userId", "userName"],
    "stateTransition": {
      "USER_LOGIN": {
        "0": [{"transitionTo": "1"}]
      }
    },
    "stateToAction": {
      "1": "welcome-nudge"
    },
    "actions": [
      {
        "actionId": "welcome-nudge",
        "type": "NUDGE_UI",
        "template": {
          "type": "BOTTOMSHEET",
          "props": {
            "title": "Welcome back, {{userName}}!"
          }
        }
      }
    ],
    "resetStates": ["1"]
  }
}

Example: User Type Segmentation​

{
  "ctaId": "user-type-cta",
  "rule": {
    "groupByConfig": {
      "groupByKeys": ["userType"],
      "maxActiveStateMachineCount": 10
    },
    "stateTransition": {
      "APP_LAUNCH": {
        "0": [
          {
            "transitionTo": "1",
            "filters": {
              "operator": "AND",
              "filter": [
                {
                  "propertyName": "userType",
                  "propertyType": "string",
                  "comparisonType": "=",
                  "comparisonValue": "premium"
                }
              ]
            }
          }
        ]
      }
    },
    "stateToAction": {
      "1": "premium-nudge"
    },
    "actions": [
      {
        "actionId": "premium-nudge",
        "type": "NUDGE_UI",
        "template": { /* ... */ }
      }
    ],
    "resetStates": ["1"]
  }
}

State Machine ID​

The state machine ID is generated from group by keys:

// If groupByKeys = ["userId"]
// State machine ID = event.userId

// If groupByKeys = ["userId", "userType"]
// State machine ID = `${event.userId}-${event.userType}`

Best Practices​

  1. Choose Relevant Keys: Group by attributes that affect Engagement behavior
  2. Set Limits: Use maxActiveStateMachineCount to prevent memory issues
  3. Use Context: Store group by keys in context for later use
  4. Test Grouping: Test with different user segments
  5. Monitor Performance: Monitor state machine count

Next Steps​