Automated Anonymous Feedback
Share
An automated feedback system using AI voice technology to collect, process, and distribute customer feedback.
Key Components:
-
Phone-Based Feedback Collection:
- Create a dedicated phone number customers can call to leave anonymous feedback
- Uses Claude (AI voice technology) to handle the conversation with callers
- Allows customers to share experiences (e.g., "I had an amazing experience with Hampton")
-
Automated Processing Pipeline:
- Feedback gets stored in a database (Airtable)
- System automatically summarizes the call content
- Processed feedback is posted to Slack for team visibility
-
Technical Stack:
- Twilio for phone number and call handling
- Airtable for database storage
- Slack for team notifications
- Claude for voice conversation
-
Implementation Example:
- Greg suggested building this for MFM (My First Million podcast)
- Sam expressed interest in creating an "MFM hotline" where people can call and leave messages
- Content could be repurposed for podcast segments
-
Current Technology Status:
- Voice AI technology is "85% there" according to Greg
- Has improved dramatically in the last six months (from "20% there" to nearly complete)
- Easy to implement ("easy peasy" to build)
46:06 - 47:19
Full video: 55:02GI
Greg Isenberg
CEO of Late Checkout and former advisor to Reddit and TikTok. Hosts The Startup Ideas Podcast, sharing insights with over 70,000 newsletter subscribers.
Interviews notable figures like Jason Fried and Eric Ries, focusing on entrepreneurship and community building.