SkillUp Transcript Analysis - 2000 Transcripts

This workflow automates the analysis of large volumes of transcript files stored in Google Drive by processing them through Claude AI for content analysis and storing the results in Airtable. It's designed to handle up to 2000 transcripts with built-in rate limiting and error handling to ensure reliable batch processing.

Purpose

No business context provided yet — add a context.md to enrich this documentation.

How It Works

1. Manual Trigger: The workflow starts when manually triggered by clicking "Test workflow"
2. List Files: Retrieves all transcript files from a Google Drive folder
3. Parallel Processing: For each transcript, the workflow simultaneously:
   • Downloads the file content from Google Drive
   • Fetches the shareable Google Drive link via API
4. AI Analysis: Sends the transcript content to Claude AI for analysis
5. Parse Results: Extracts structured data from Claude's response
6. Store Results: Creates a new record in Airtable with the analysis results
7. Rate Limiting: Waits 1 second between requests to respect API limits
8. Status Logging: Logs processing status for monitoring
9. Error Handling: Continues processing even if individual transcripts fail

Workflow Diagram

graph TD
    A[Manual Trigger] --> B[List Transcripts from Google Drive]
    B --> C[Read Transcript Content]
    B --> D[Get Google Drive Link]
    C --> E[Analyze Transcript with Claude]
    E --> F[Parse Claude Response]
    F --> G[Create Airtable Record]
    G --> H[Rate Limiter - 1 sec]
    H --> I[Log Processing Status]
    J[Error Handler] --> I

Trigger

Manual Trigger: This workflow is started manually by clicking the "Test workflow" button. This is appropriate for batch processing operations that don't need to run on a schedule.

Nodes Used

Node Type	Node Name	Purpose
Manual Trigger	When clicking 'Test workflow'	Starts the workflow manually
Google Drive	List Transcripts from Google Drive	Retrieves list of transcript files
Google Drive	Read Transcript Content	Downloads file content for analysis
HTTP Request	Get Google Drive Link	Fetches shareable link for each file
OpenAI Message	Analyze Transcript with Claude	Processes transcript through Claude AI
JSON Extract	Parse Claude Response	Extracts structured data from AI response
Airtable	Create Airtable Record	Stores analysis results in database
Wait	Rate Limiter - 1 sec between requests	Prevents API rate limit violations
Set	Log Processing Status	Records processing status for monitoring
Error Trigger	Error Handler - Continue on Failure	Handles errors without stopping workflow

External Services & Credentials Required

Google Drive

• Purpose: File storage and retrieval
• Credentials: Google OAuth2 or Service Account
• Permissions: Read access to transcript folder

Google API

• Purpose: Fetching file metadata and links
• Credentials: Google API credentials
• Scopes: Drive API access

Claude AI (via OpenAI Message node)

• Purpose: Transcript content analysis
• Credentials: Anthropic API key
• Note: Despite using OpenAI Message node, this connects to Claude

Airtable

• Purpose: Storing analysis results
• Credentials: Airtable API key
• Requirements: Access to specified base and "Transcripts" table

Environment Variables

No specific environment variables are defined in this workflow. All configuration is handled through node parameters and credentials.

Data Flow

Input

• Transcript files stored in Google Drive (various formats supported by Google Drive API)
• Files are processed in batches as returned by the Drive API

Processing

• Each transcript is analyzed by Claude AI for content insights
• Results are parsed into structured JSON format
• Google Drive links are preserved for reference

Output

• Airtable records containing:
  • Original transcript metadata
  • Claude's analysis results
  • Google Drive shareable links
  • Processing timestamps

Error Handling

The workflow includes robust error handling:

• Error Trigger Node: Catches failures and allows the workflow to continue processing remaining transcripts
• Continue on Failure: Individual transcript failures don't stop the entire batch
• Rate Limiting: 1-second delays prevent API throttling errors
• Status Logging: Tracks successful and failed processing attempts

Known Limitations

• Manual trigger only - no automated scheduling
• Rate limited to 1 request per second (may be slow for 2000 transcripts)
• No retry mechanism for failed individual transcripts
• Dependent on external API availability (Google Drive, Claude, Airtable)

Related Workflows

No related workflows specified in the current context.

Setup Instructions

1. Import Workflow

   • Copy the workflow JSON
   • Import into your n8n instance

2. Configure Credentials

   • Set up Google Drive OAuth2 or Service Account credentials
   • Add Google API credentials with Drive API access
   • Configure Anthropic API key for Claude access
   • Set up Airtable API credentials

3. Configure Nodes

   • List Transcripts: Specify the Google Drive folder containing transcripts
   • Airtable: Set the correct base URL and table name ("Transcripts")
   • Claude Analysis: Configure the AI prompt for your specific analysis needs

4. Test Setup

   • Start with a small subset of transcripts
   • Verify all credentials are working
   • Check that Airtable records are created correctly

5. Production Run

   • Activate the workflow
   • Click "Test workflow" to process all 2000 transcripts
   • Monitor the execution logs for any errors

6. Post-Processing

   • Review Airtable for complete results
   • Check error logs for any failed transcripts
   • Re-run individual failures if needed