​
The Web Dashboard

The easiest way to get started with ElevenLabs Conversational AI is through our web dashboard.

You can access your dashboard here. The web dashboard enables you to:

  • Create and manage AI assistants
  • Configure voice settings and conversation parameters
  • Review conversation analytics and transcripts
  • Manage API keys and integration settings

The web dashboard uses our Web SDK (link to it) under the hood to handle real-time conversations.

​
Pierogi Palace Assistant

In this guide, we’ll create an AI assistant for “Pierogi Palace” - a modern Polish restaurant that takes orders through voice conversations. Our assistant will help customers order traditional Polish dishes with a contemporary twist.

The Pierogi Palace menu showcases traditional Polish dishes that customers can order through the AI assistant

The assistant will guide customers through:

  • Menu Selection

    • Various pierogi options with traditional and modern fillings
    • Portion sizes (available in dozens)

  • Order Details

    • Quantity confirmation
    • Price calculation in Polish złoty
    • Order review and modifications if needed

  • Delivery Information

    • Delivery address collection
    • Estimated preparation time (10 minutes)
    • Delivery time calculation based on location
    • Payment method confirmation (cash on delivery)

​
Assistant Setup

In this guide, we’ll walk through configuring your Pierogi Palace assistant using ElevenLabs Conversational AI. We’ll set up the assistant’s voice, language model, and transcription settings to help customers place orders seamlessly.

​
Prerequisites

​
1. Access the Dashboard

1

Sign In to ElevenLabs

Go to elevenlabs.io and sign in to your account.

2

Navigate to Conversational AI

In the ElevenLabs dashboard, click on Conversational AI in the left sidebar.

Navigating to the Conversational AI section

​
2. Create Your Assistant

1

Start Creating a New Assistant

  • Click the + button to create a new AI Agent.
  • Choose the Blank Template option & call the agent Pierogi Palace.

Creating a new assistant

2

Configure Assistant Details

Set the First message & System prompt fields to the following, leaving the Knowledge Base and Tools empty for now:

​
3. Configure Voice Settings

1

Select a Voice for Your Assistant

In this step, you can choose from over 3,000 life-like voices available in ElevenLabs, for this demo we will be using Jessica’s voice.

Assistant voice configuration

Higher quality settings may increase response time slightly. For an optimal customer experience, we recommend balancing quality and latency based on your assistant’s expected use case.

​
4. Test Your Assistant

1

Converse with Your Assistant

  • Press the Order button and try ordering some Pierogi to see how the assistant handles the conversation.

Testing your assistant

​
5. Configure Data Collection

Configure evaluation criteria and data collection to analyze conversations and improve your assistant’s performance.

1

Configure Evaluation Criteria

Navigate to the ANALYSIS section in your assistant’s settings to define custom criteria for evaluating conversations.

  1. Goal Prompt Criteria This passes the conversation transcript to the LLM to verify if specific goals were met. Results will be:
  • success
  • failure
  • unknown

Plus a rationale explaining the chosen result.

Configure the following fields:

  • Name: Enter a descriptive name
  • Prompt: Enter detailed instructions for evaluating the conversation
Evaluation Examples
1. Name: order_completion
   Prompt: Evaluate if the conversation resulted in a successful order completion.
   Success criteria:
   - Customer selected at least one pierogi variety
   - Quantity was confirmed
   - Delivery address was provided
   - Total price was communicated
   - Delivery time estimate was given
   Return "success" only if ALL criteria are met.

2. Name: customer_satisfaction
   Prompt: Analyze the conversation for signs of customer satisfaction.
   Success criteria:
   - Customer's questions were answered clearly
   - No repeated requests for clarification
   - No signs of frustration or confusion
   - Positive or neutral customer responses
   Return "failure" if there are clear signs of dissatisfaction.

3. Name: menu_explanation
   Prompt: Evaluate if the assistant properly explained menu options when asked.
   Success criteria:
   - Mentioned available pierogi varieties
   - Provided prices when relevant
   - Explained portion sizes (dozens)
   Return "unknown" if menu items were not discussed.
2

Set Up Data Collection

In the Data collection section, define specifications for extracting data from conversation transcripts.

Click Add data collection item and configure:

  • Data type: Select “string”
  • Identifier: Enter a unique identifier for this data point
  • Description: Provide detailed instructions for the LLM about how to extract the specific data from the transcript

Example data collection items:

Data Collection Examples
1. Identifier: order_type
   Description: Extract the type of order from the conversation. Should be one of: "delivery", "pickup", "inquiry_only"

2. Identifier: ordered_items
   Description: List all pierogi varieties and quantities ordered in the format: "item: quantity". If no order was placed, return "none"

3. Identifier: delivery_zone
   Description: Based on the delivery address, categorize the location as: "central_zakopane", "outer_zakopane", or "outside_delivery_zone"

4. Identifier: interaction_type
   Description: Categorize the conversation as: "completed_order", "abandoned_order", "menu_inquiry", or "general_inquiry"
3

View Conversation History

You can view evaluation results and collected data for each conversation in the history tab.

Your Pierogi Palace assistant is now ready to take orders 🥟! The assistant can handle menu inquiries, process orders, and provide delivery estimates.

Was this page helpful?

IntroductionNext.JS