The meeting provided a detailed walkthrough of automating outbound calls using NA10 workflows integrated with Google Calendar and Retail AI for voice AI-powered appointment confirmations.
The demonstration covered setting up triggers, connecting APIs, configuring AI agents, and enhancing contextual knowledge for more effective outreach.
Best practices, potential improvements, compliance considerations, and available resources (AI Success Lab and AI Profit Boardroom) were highlighted for further support.
Action Items
N/A: No specific dated action items or owners referenced in this transcript.
Outbound Call Automation Overview
Demonstrated how to build an outbound call automation workflow using NA10 and Retail AI.
Workflow includes fetching new appointments from Google Calendar, processing them with an AI agent, and triggering automated phone calls to leads or clients.
Emphasized flexibility: can be used for sales, appointment confirmations, or lead generation.
Stressed the importance of legal compliance when using AI for outbound calls.
Setting Up the Workflow
Step-by-step setup:
Schedule workflow trigger (e.g., daily at 9 a.m.).
Connect and authenticate Google Calendar to obtain upcoming appointment data.
Pass appointment data to an AI agent via NA10 using OpenAI or similar providers.
Configure AI agent prompt with contextual event and participant information, system messages, and structured output parsing.
Actionable templates and prompts are available via the AI Success Lab for easier replication and customization.
Configuring Retail AI and Integrating with NA10
Retail AI handles the outbound calling with voice AI; requires creating a single prompt agent, setting a welcome message, and optionally building a multilingual knowledge base.
Users can select from various LLM options and voice types depending on cost and use case.
Must purchase ($2, using free credits) and verify a phone number in Retail AI to activate outbound calls.
Retail AI's API is linked to NA10 via an HTTP POST request, including relevant JSON payloads for call context.
Enhancing Agent Intelligence
Demonstrated adding knowledge base content to Retail AI agents using web pages, files, or call transcripts.
Auto-sync features keep the knowledge base up-to-date with website changes or other resources.
Customizing prompts to limit agent verbosity and focus on confirming appointments or asking targeted questions is recommended for better customer experience.
Testing and Troubleshooting
Each workflow stage should be tested sequentially to ensure correct data handoff and prevent errors.
Issues can be debugged at each step (e.g., parsing event data, API responses, or call initiation).
Available Resources & Support
Free templates, instructions, and resources are accessible via the AI Success Lab.
Paid support, coaching, and a peer community are available through the AI Profit Boardroom for users needing more hands-on guidance.
Decisions
Recommended best practices for AI outbound call scripting — Limit agent prompts to one question at a time for clarity and improved user experience.
Open Questions / Follow-Ups
Compliance: Clarification needed on legal requirements for automated AI outbound calling in various jurisdictions; users encouraged to conduct their own research.