AI Interaction Data Capture in Talkative

AI Interaction Data Capture is a way for Talkative to automatically classify and assign interaction data values. Interaction Data is metadata attached to conversations, consisting of user-defined key-value pairs. Typical values include “Name” and “Email” but can also include “Category,” “Account Number,” or anything else you desire.

Overview

Talkative’s AI Interaction Data Capture uses natural language processing to read conversation transcripts and extract predefined data fields (e.g., name, policy number, date of birth, summary). The system dynamically assigns these values as metadata to conversations. Key benefits include:

• Natural Conversation Flow: No need for rigid scripted questions; data is captured seamlessly in the background.
• Real-Time Accuracy: Updates occur with every new message, ensuring agents always have up-to-date information.
• Enhanced Data Quality: Standardized key-value pairs can later be used with custom analytics, CRM integration, and more.

How It Works

1. Configuration Setup:

   • Create a Config: Define the data fields you wish to capture (e.g., Name, Policy Number, Date of Birth, Summary).
   • Define Prompts: Provide a natural language description of what each data point should look like. This acts as a prompt to the AI to categorize the data from the transcript.
   • Queue Mapping: Apply your configuration to a specific queue. Each queue can have one configuration, but you can create multiple configurations overall.

2. Data Capture Process:

   • Real-Time Extraction: As the conversation unfolds—via voice calls, video chats, or messaging—the AI listens to the transcript and extracts key information.
   • Dynamic Updates: After each message, the system updates the interaction data in real time.
   • Post-Interaction: Captured data is automatically sent to your CRM or data warehouse and can be reviewed or adjusted in post-interaction wrap-up forms.

3. Agent Experience:

   • Live Console View: Agents can view the Interaction Data in the console under the information tab and in the Interaction Log page.
   • Wrap-Up Forms: Agents can check or override the captured data during their wrap-up process, ensuring accuracy before finalizing records.

Instructions to Use AI Interaction Data Capture

This feature is available on all Talkative subscriptions. You must be an Account Holder to configure it.

Step 1: Access the Template List View

• Search for “Capture” in your Talkative settings.

Step 2: Create a Config

• Create a New Configuration: You can create multiple configurations and map them to different queues. Remember, you can only map one configuration to a specific queue, but there is no limit to the number of configurations you can create.

Step 3: Create Values and Assign to Queues

• Choose the Queue(s): Select the queue(s) you want the configuration to work for.
• Create Interaction Data Fields: Think of the Interaction Data description as a prompt for the AI to categorize the data based on the transcript.
• 

Step 4: Watch It Work

• Live Data Updates: Agents can view the Interaction Data in the console under the Information tab.
• Interaction Log: All collected data is visible on the Interaction Log page, updating with every new message.
• Video Chat Support: The system works equally well with video chat interactions.

Note: This feature involves sending your transcripts to OpenAI for analysis and data classification.

Example Walkthrough

Consider a typical voice call:

1. Before the Call:

   • A configuration is set up to capture fields such as “Name,” “Policy Number,” “Date of Birth,” and “Summary.”
   • The configuration is applied to the call queue.

2. During the Call:

   • As the customer provides their information (e.g., “My name is John D. Smith” or “My policy number is 123456”), the AI listens to the transcript.
   • The system extracts and updates key data points after every message, such as capturing “20 October 1991” as the customer’s date of birth, formatted according to your configuration (e.g., using British date format).

3. After the Call:

   • The agent sees a complete set of data fields populated in the interaction log.
   • The agent can verify or adjust any captured details in a wrap-up form.
   • Once confirmed, the finalized interaction data is automatically sent to your CRM or data warehouse.

Tips and Suggested Data Fields

Here are some data fields you might want to try:

• Customer Status: New vs. Returning Customer, Loyalty Level (e.g., Silver, Gold, VIP).
• Customer Satisfaction/Outcome: Resolved, Unresolved, Escalated, Follow-up Needed, Abandoned.
• Language Used
• Product/Service Mentioned
• Request Type: Information Request, Issue Resolution, Order Placement, Product Feedback.
• Follow Up Required?
• Complaint (Yes/No)?

You can also use Analytics custom reports and interaction log search filters to retrieve interactions with specific AI data types.

Benefits

• Efficient Data Collection: Automatically captures key details without interrupting the conversation flow.
• Accurate and Real-Time: Instant updates provide agents with timely, verified information.
• Customizable: Create and assign multiple configurations across various queues based on your needs.
• Integrated Workflow: Seamlessly integrates with CRM systems and supports post-interaction wrap-up verification.

Getting Help

If you have any questions, encounter issues, or have feedback on this feature, please contact our support team. We’re here to help you get the most out of Talkative’s AI Interaction Data Capture capabilities.