How to Create a Custom Workflow in Dialora

Workflows let you customize how your AI voice agent handles conversations from what it says to how it gathers data, routes calls, or interacts with external tools.

Step 1: Go to the Workflow Section

Log in to your Dialora dashboard.
On the left panel, scroll until you see Workflows and click on it.
Click the “Create Workflow” button in the center of the screen.

Step 2: Start Building Your Workflow

You’ll see a Start node on the screen.

Click on the + (plus) sign under Start to begin adding steps.

Step 3: Choose What Happens in the Workflow

A menu will appear with the following action blocks:

Say – The AI speaks something.
Gather – Ask a question and capture user responses.
Condition – Branch logic based on responses (e.g., Yes → path A, No → path B).
API Request – Trigger an API call based on user input or logic.
Transfer Call – Hand off the call to a human agent if needed.

Example: Using the “Say/Speak” Block

Click on Say, then click Edit.
Enter the message or prompt you want the AI to say at that point in the call.

Example: Using “Condition” for Branching

Use Condition if you need different actions based on what the user says.
Example: -If the user says Yes → Route to Confirmation Block. -If the user says No → Route to Follow-Up Block.

Example: Using “Gather” to Collect Info

Use Gather when you want to ask the caller for specific info (like name, email, service type).
Input your question(s), and the AI will collect responses during the call.

Example: Using “API Request” for Integration

Use API Request when you want the workflow to interact with other tools (like a CRM or booking system).
Depending on the user’s response, the AI can make a custom API call (e.g., book an appointment, fetch data).

Example: Using “Transfer Call” to Escalate

If the call requires human support, add a Transfer Call block at any stage.
Input the phone number and country code where you want to route the call.

Pro Tip:

You can mix and match blocks to create smart, dynamic conversations. Start simple, test it live, and then layer in logic, data capture, and integrations to make your voice agent more powerful.