Skip to main content

What is a Flow?

A flow is an automation workflow that connects triggers (events that start the flow) with actions (tasks that execute when triggered). Flows help you automate post-call processing, integrate with external systems, and enhance your AI calling capabilities.

Understanding Flow Components

Trigger: The Starting Point

The trigger is the event that starts your flow. It determines when and how often the flow executes. Available Trigger Types:
  • Call Ended Trigger: Activates when an AI call completes
  • Inbound Call Variable Injection: Triggers before the AI assistant picks up an incoming call
  • Schedule Trigger: Runs at specific times or intervals
  • Webhook Trigger: Activates when an external system sends a webhook
  • Event Trigger: Responds to specific service events
Each trigger provides different data and context. For example, the “Call Ended” trigger gives you access to the full conversation transcript, call duration, and outcome.

Action: What Happens Next

Actions define what occurs when the flow is activated. They control the automation tasks that execute after the trigger fires. Common Action Types:
  • CRM Updates: Update HubSpot, Salesforce, or other CRM systems
  • Data Storage: Save call data to Google Sheets, databases, or files
  • Notifications: Send emails, SMS, or Slack messages
  • Webhooks: Trigger external APIs or custom integrations
  • Campaign Management: Add leads to campaigns or update lead status
  • Code Execution: Run custom JavaScript code for complex logic

Real-World Flow Example

Visual diagram showing flow components: trigger on the left connected to actions on the right Example: Post-Call CRM Update Flow
  1. Trigger: Call Ended
  2. Action 1: Extract key information from transcript (e.g., customer name, inquiry type)
  3. Action 2: Update HubSpot contact with call outcome
  4. Action 3: Send email notification to sales team if lead is qualified

Best Practices for Building Flows

Start Simple

Begin with a single trigger and one action. Test thoroughly before adding complexity.

Use Variables Effectively

Leverage data from triggers (like call transcripts or lead information) to personalize actions.

Handle Errors

Add error handling for actions that might fail (API timeouts, invalid data, etc.).

Test Your Flows

Use the debugging tools to test flows before deploying them to production.
Next Steps