Willingness to Pay

An interactive web application that helps users analyze call transcripts to assess customer willingness to pay.

Features

• Analyzes call transcripts to assess customer willingness to pay
• Customizable system prompts and analysis criteria
• Dark/light theme support
• Interactive results table with detailed modal view
• Keyboard navigation support
• Secure authentication via LLM Foundry API
• Supports multiple transcript analysis in batch
• JSON schema validation for consistent responses

Usage

1. Log in using your LLM Foundry credentials
2. Either use the default transcripts or enter your own:
   • Add transcripts separated by ==========
   • Customize the system prompt if needed
   • Modify the analysis criteria (one per line)
3. Click "Analyze" to process the transcripts
4. View results in the interactive table:
   • ✅ indicates positive responses
   • ❌ indicates negative responses
   • Click any row to view detailed analysis
   • Use ↑/↓ keys to navigate between results

Screenshot

Installation

Prerequisites

Local Setup

1. Clone this repository:

git clone https://github.com/gramener/willingnesstopay.git
cd willingnesstopay

2. Serve the files using any static web server. For example, using Python:

python -m http.server

3. Open http://localhost:8000 in your web browser

Deployment

On Cloudflare DNS, proxy CNAME willingnesstopay.straive.app to gramener.github.io.

On this repository's page settings, set

• Source: Deploy from a branch
• Branch: main
• Folder: /

Technical Details

Architecture

The application follows a simple single-page architecture:

• Frontend-only implementation using vanilla JavaScript and ESM modules
• Streaming LLM responses for real-time analysis feedback
• Bootstrap for responsive UI components
• lit-html for efficient DOM updates
• JSON schema validation for API responses

Dependencies

• lit-html: Template rendering
• asyncLLM: Streaming LLM responses
• partial-json: JSON parsing
• Bootstrap: UI framework
• Bootstrap Icons: Icon set

The application uses the LLM Foundry API for:

• Authentication via token-based access
• GPT-4 powered transcript analysis
• Streaming response handling

Development

Project Structure

├── index.html # Main HTML file
├── script.js # Main application logic
├── style.css # Styling
└── README.md # Documentation

License

MIT